Automobile mechanical fault prediction and early warning method

Abstract

This invention discloses a method for predicting and warning of automotive mechanical faults, comprising: acquiring vibration and sound data of vulnerable vehicle components under different vehicle conditions, and establishing a historical database of vulnerable vehicle components under different vehicle conditions; importing the historical database of vulnerable components into a robot learning model, and determining the vehicle status based on the vibration and sound data; extracting vibration and sound characteristics of each vulnerable component under abnormal vehicle conditions, under different vehicle conditions and at different fault stages, and drawing abnormal noise fault maps of different vulnerable components under different vehicle conditions based on the extracted feature data; acquiring real-time vibration and sound data of each vulnerable component of the vehicle, corresponding the acquired real-time vibration and sound data with the abnormal noise fault maps, and determining the cause and location of the vehicle fault; determining the fault level based on the cause and location of the vehicle fault, and issuing a fault warning to the vehicle or user terminal.

Description

Technical Field

[0001] This invention relates to the field of automotive fault diagnosis, and in particular to a method for predicting and warning of automotive mechanical faults. Background Technology

[0002] Currently, to ensure vehicle safety, sensor networks built into various electrical subsystems (such as battery management systems and motor controllers) are typically used to monitor physical quantities such as voltage, current, temperature, and pressure, thereby predicting failures of electrical or thermal management components. However, vehicles contain numerous critical mechanical components, such as ball joints, connecting rods, and bushings in the suspension system; tie rods and universal joints in the steering system; calipers and discs in the braking system; and various bearings and drive shafts. Before functional failure, these components often undergo a physical degradation process due to wear, loosening, fatigue cracks, or poor lubrication. A significant characteristic of this process is the generation of characteristic abnormal vibrations and noises (i.e., "abnormal sounds"). Due to cost, space, and reliability considerations, it is impossible to install vibration or displacement sensors at every mechanical point.

[0003] Diagnosing mechanical noises primarily relies on the driver's subjective report or the mechanic's experience during maintenance. This method has significant drawbacks: 1) Delay: The noise is often only addressed when it becomes very noticeable and attracts the driver's attention, at which point the component may already be severely worn, posing a safety hazard. 2) Subjectivity: Different drivers have vastly different sensitivities to sound and can describe it in different ways, hindering accurate diagnosis. 3) Lack of early warning: It cannot provide early warning of potential faults, contradicting the principles of predictive maintenance.

[0004] To address the aforementioned issues, Chinese Patent Application No. 2022103728829 discloses a method for automatically analyzing the causes of abnormal noises in vehicles. This method includes acquiring unknown abnormal noise data collected inside the vehicle during operation, extracting identifiable features from the unknown abnormal noise data, and analyzing these identifiable features using a trained machine learning model to determine the cause of the abnormal noise. The trained machine learning model is trained using known abnormal noise data with labeled causes. Analyzing the causes of vehicle abnormal noises can include not only identifying the type and source of the noise, but also further determining the underlying cause. While this method can determine the cause of abnormal noises after training with the noise data, it cannot precisely pinpoint the exact location of the fault. This method of judging abnormal noise faults solely based on sound lacks tactile information (such as vibration) from mechanical components, making the cause judgment too simplistic. Furthermore, the audio information is subject to environmental noise interference, resulting in inaccurate fault judgments and potential misjudgments. Additionally, this method cannot provide early warnings or determine the fault level. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the existing technology, the purpose of the present invention is to solve the problems of the existing vehicle abnormal noise fault diagnosis, which are limited to single fault analysis causes, single data collection, and inability to provide early warning and judgment of fault level. The present invention provides a method for predicting and warning of automotive mechanical faults, which can collect multi-source data on the causes of vehicle faults, determine the fault location, and accurately determine the fault cause and fault level.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: a method for predicting and warning of automotive mechanical faults, comprising the following steps: S1, acquiring vibration and sound data of vulnerable vehicle components under different vehicle conditions, and establishing a historical database of vulnerable vehicle components under different vehicle conditions; S2, importing the historical database of vulnerable components into a robot learning model, and judging the vehicle state based on the vibration and sound data of each vulnerable component under different vehicle conditions, wherein the vehicle state includes normal state and abnormal state; S3, extracting the vibration and sound characteristics of each vulnerable component under abnormal vehicle conditions under different vehicle conditions and different fault stages, and drawing abnormal noise fault maps of different vulnerable components under different vehicle conditions based on the extracted feature data; S4, acquiring real-time vibration and sound data of each vulnerable component of the vehicle, corresponding the acquired real-time vibration and sound data with the abnormal noise fault maps, and determining the cause and location of the vehicle fault; S5, determining the fault level based on the cause and location of the vehicle fault, and issuing a fault warning to the vehicle or user terminal. In the initial stage of fault prediction for vulnerable vehicle components, vibration and sound data of these components under different vehicle conditions were collected, both in normal and abnormal (fault) states. This ensures that during deep learning, each vulnerable component has corresponding vibration and sound data under different conditions, resulting in richer data for deep learning. This allows for assessment of the vehicle's state and mechanical faults from multi-dimensional data, effectively ensuring more accurate fault diagnosis. Specifically, fault diagnosis learning utilizes comprehensive data from vibration, driving conditions, and sound data to ensure more accurate fault cause identification. Furthermore, the source of the fault sound and vibration data determines the approximate location of the vulnerable component. Based on the common fault types at that location and the corresponding sound and vibration data, the cause of the fault can be clearly determined, avoiding misdiagnosis. Finally, by combining the fault cause, the source of the fault sound and vibration data, and the current driving conditions with vibration and audio characteristics, the specific location of the fault can be determined, along with the fault level and a warning issued. This fault prediction and early warning method can effectively issue warnings as soon as a fault occurs, and remind drivers to perform maintenance when the fault is minor.

[0007] Furthermore, the vibration data is obtained through vibration sensors installed on various vulnerable vehicle components, and the sound data is obtained using microphone arrays installed on these components. The vibration data includes vibration frequency, vibration duration, and vibration onset time, while the sound data includes audio frequency, phase, pitch, timbre, spectrum, and duration. Sound data obtained through microphone arrays is more comprehensive, allowing for the determination of the specific location of the abnormal noise based on the phase of the microphones in the array, combined with audio and pitch data. The vibration frequency in the vibration data reflects the degree of wear and looseness of mechanical components, and the vibration onset time, combined with vehicle operating conditions, can determine the timing of the abnormal noise. Furthermore, by combining the timing of the abnormal noise with the source of the sound data, the specific fault location can be determined.

[0008] Furthermore, the vibration and sound data of the vehicle's vulnerable components under different vehicle conditions are obtained at different driving speeds, loads, road conditions, and driving states. The road conditions include flat roads, downhill sections, and uphill sections, and the driving states include going straight, turning left, and turning right. The same fault, but with different fault levels, will correspond to different sound and vibration data under different driving speeds, loads, road conditions, and driving states. Therefore, collecting vibration and sound data at different driving speeds, loads, road conditions, and driving states provides more data and information for subsequent learning and judgment, leading to more accurate fault diagnosis.

[0009] Furthermore, the abnormal noise fault map includes the audio, phase, pitch, timbre, spectrum, and duration of abnormal noises corresponding to the cause of the fault in different operating conditions when different damaged components occur in the vehicle, as well as the real-time vibration frequency, vibration duration, and vibration onset time corresponding to the cause of the fault under the same operating conditions. The abnormal noise fault map contains detailed parameters and comprehensive data with multiple parameters and dimensions. After being correlated with real-time data, the corresponding fault can be determined, ensuring more accurate subsequent fault diagnosis.

[0010] Furthermore, the fault level is determined based on the abnormal noise vibration frequency, vibration duration, abnormal noise pitch, and abnormal noise duration, including Level 1, Level 2, and Level 3 faults. Level 1 faults are minor faults that will not affect driving safety; Level 2 faults are low-level safety faults that require prompt repair; and Level 3 faults are high-level faults that require immediate repair. The determination and classification of fault levels provides drivers with a basis for maintenance, preventing drivers from aggravating faults or causing safety accidents due to negligence.

[0011] Furthermore, the location of the abnormal noise fault is determined by the vehicle condition, phase, pitch, and source of the sound when the abnormal noise occurs. Determining the location of the abnormal noise fault based on the vehicle condition, sound phase, pitch, and source of the sound allows for precise location down to the specific faulty component. Attached Figure Description

[0012] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0013] Figure 1 This is a flowchart of the automotive mechanical fault prediction and early warning method in the embodiment. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0016] It should be noted that similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the figures, or the orientation or positional relationship commonly used when the product is in use. They are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance. In addition, the terms "horizontal," "vertical," etc., do not indicate that the component is required to be absolutely horizontal or suspended, but can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted. In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0017] Example: See Figure 1The automotive mechanical fault prediction and early warning method provided in this embodiment includes the following steps: S1, acquiring vibration and sound data of vulnerable vehicle components under different vehicle conditions, and establishing a historical database of vulnerable vehicle components under different vehicle conditions; S2, importing the historical database of vulnerable components into a robot learning model, and judging the vehicle status based on the vibration and sound data of each vulnerable component under different vehicle conditions, wherein the vehicle status includes normal status and abnormal status; S3, extracting the vibration and sound characteristics of each vulnerable component under abnormal vehicle conditions under different vehicle conditions and different fault stages, and drawing abnormal noise fault maps of different vulnerable components under different vehicle conditions based on the extracted feature data; S4, acquiring real-time vibration and sound data of each vulnerable component of the vehicle, corresponding the acquired real-time vibration and sound data with the abnormal noise fault map, and determining the cause and location of the vehicle fault; S5, determining the fault level based on the cause and location of the vehicle fault, and issuing a fault warning to the vehicle or user terminal. In the initial stage of fault prediction for vulnerable vehicle components, vibration and sound data of these components under different vehicle conditions were collected, both in normal and abnormal (fault) states. This ensures that during deep learning, each vulnerable component has corresponding vibration and sound data under different conditions, resulting in richer data for deep learning. This allows for assessment of the vehicle's state and mechanical faults from multi-dimensional data, effectively ensuring more accurate fault diagnosis. Specifically, fault diagnosis learning utilizes comprehensive data from vibration, driving conditions, and sound data to ensure more accurate fault cause identification. Furthermore, the source of the fault sound and vibration data determines the approximate location of the vulnerable component. Based on the common fault types at that location and the corresponding sound and vibration data, the cause of the fault can be clearly determined, avoiding misdiagnosis. Finally, by combining the fault cause, the source of the fault sound and vibration data, and the current driving conditions with vibration and audio characteristics, the specific location of the fault can be determined, along with the fault level and a warning issued. This fault prediction and early warning method can effectively issue warnings as soon as a fault occurs, and remind drivers to perform maintenance when the fault is minor.

[0018] The vulnerable vehicle components described in this embodiment include engine valve mechanisms, piston connecting rod assemblies, transmission components, shock absorbers, brake discs, steering systems, wheels, and other mechanical connecting parts. Based on this, this embodiment places at least one waterproof and dustproof microphone inside each of the four wheel arches to collect abnormal noises from the suspension and braking systems; two to three microphones are placed in the center of the chassis to collect overall chassis noises; and existing vehicle-mounted microphone arrays are used to assist in analysis within the passenger compartment.

[0019] In this embodiment, vibration data is obtained through vibration sensors installed on various vulnerable vehicle components, while sound data is obtained using microphone arrays installed on these components. The vibration data includes vibration frequency, vibration duration, and vibration onset time, while the sound data includes audio frequency, phase, pitch, timbre, spectrum, and duration. Sound data obtained through microphone arrays is more comprehensive, allowing for the determination of the specific location of abnormal noises based on the phase of the microphones in the array, combined with audio and pitch data. The vibration frequency in the vibration data reflects the degree of wear and looseness of mechanical components, and the vibration onset time, combined with vehicle operating conditions, can determine the timing of the abnormal noise. Furthermore, by combining the timing of the abnormal noise with the source of the sound data, the specific location of the fault can be identified.

[0020] Under different driving speeds, loads, road conditions, and driving states, the sound and vibration data corresponding to the same fault and different fault levels will also be different (for example, abnormal noises are more obvious during rapid acceleration or climbing, abnormal noises during acceleration / deceleration may be related to the transmission system, and abnormal noises during turning require checking the steering or suspension components). To avoid misdiagnosis of abnormal noise faults, the vibration and sound data of the vehicle's vulnerable components under different driving conditions are obtained from vibration and sound data of the vehicle's vulnerable components under different driving speeds, loads, road conditions, and driving states. The road conditions include flat roads, downhill sections, and uphill sections, and the driving states include straight driving, left turns, and right turns. Collecting vibration and sound data under different driving speeds, loads, road conditions, and driving states provides more basis and sources for judgment in subsequent learning and judgment, and the corresponding judgment of fault causes is more accurate.

[0021] In this embodiment, the abnormal noise fault map shows the audio, phase, pitch, timbre, spectrum, and duration of abnormal noises corresponding to the cause of the fault in different operating conditions when different damaged components occur in the vehicle, as well as the real-time vibration frequency, vibration duration, and vibration onset time corresponding to the cause of the fault under the same operating conditions. The abnormal noise fault map contains detailed parameters and comprehensive data with multiple parameters and dimensions. After being correlated with real-time data, the corresponding fault can be determined, ensuring more accurate subsequent fault diagnosis.

[0022] Furthermore, the fault level is determined based on the abnormal noise vibration frequency, vibration duration, abnormal noise pitch, and abnormal noise duration, including Level 1, Level 2, and Level 3 faults. Level 1 faults are minor faults (such as loose trim pieces or slight chassis noise), which do not affect driving safety. When a fault warning is issued at this level, it is recommended that the driver have the vehicle inspected during the next maintenance check. Level 2 faults are low-level safety faults requiring prompt repair (such as a possible loose left front suspension). Level 3 faults are high-level faults (such as severe abnormal noise from the right rear wheel bearing), requiring immediate repair. The judgment and classification of fault levels provide drivers with a basis for maintenance, preventing drivers from aggravating faults or causing safety accidents due to negligence.

[0023] Furthermore, the location of the abnormal noise fault is determined by the vehicle condition, phase, pitch, and source of the sound when the abnormal noise occurs. Determining the location of the abnormal noise fault based on the vehicle condition, sound phase, pitch, and source of the sound allows for precise location down to the specific faulty component.

[0024] If the current real-time vehicle status is that the vehicle is turning, and the collected abnormal noise data comes from the right front wheel, and the collected noise is a regular "clicking" sound with a low volume that only occurs when turning, then according to the information corresponding to the abnormal noise fault spectrum, it is determined that "slight wear of the outer CV joint of the right front wheel drive half shaft" is a level two fault, and a warning is issued, recommending that the user check and repair it.

[0025] To facilitate understanding, taking the Changan Oushang X5 1.5T turbocharged passenger vehicle as an example, historical data was first collected for different vulnerable components of this model under different operating conditions. Then, using the method described above, abnormal noise fault maps for different vulnerable components under different vehicle conditions were established. Based on these abnormal noise fault maps, the method for real-time prediction and fault warning is as follows.

[0026] Example 1:

[0027] Data acquisition: Two piezoelectric vibration sensors are fixed on the intake side end cap of the engine block and the connection position between the intake manifold and the cylinder block, respectively, to collect data such as vibration frequency and vibration duration; two waterproof and dustproof microphones are arranged on the intake side in the engine compartment, and one vehicle microphone is arranged on the left side of the center console in the passenger compartment to collect audio, spectrum, tone and other sound data.

[0028] The actual vehicle conditions were as follows: The vehicle was idling, unloaded, stationary on a flat road, with an engine coolant temperature of 85℃ and an idle speed of 750±50 r / min. Vibration data: A continuous vibration frequency of 280±15Hz was detected on the intake side of the engine block. The vibration started synchronously with the engine idling start, and the vibration duration was continuous and uninterrupted with the idling speed. The vibration amplitude was 0.35mm / s. Sound data: A continuous "ticking" sound was collected in the 270-290Hz audio range. The tone was sharp, and a clear characteristic peak appeared at 280Hz in the spectrum. The timbre was a metallic knocking sound, and the duration of the sound was consistent with the engine idling time, without interruption.

[0029] Fault matching and location involves comparing real-time collected vibration and sound data with the fault spectrum database of abnormal noises in Changan Oushang series passenger vehicles. A vibration frequency of around 280Hz combined with a sharp, metallic knocking "tap-tap" sound is a typical characteristic of excessive intake valve clearance in the engine. By combining the vibration sensor's collection location (intake side of the engine block) and the sound source (intake manifold area), the fault location is accurately located as excessive intake valve clearance in the third cylinder of the engine intake valve assembly.

[0030] Fault Level Determination: According to the established fault level classification standard, this fault is a Level 2 fault (low-level safety fault). The fault only affects the smoothness of engine operation and does not affect driving safety at present. However, long-term operation will aggravate the wear of valves and valve seats, leading to more serious mechanical failures.

[0031] Warning output: Send a fault warning to the vehicle-mounted infotainment system and the user's mobile app. The warning information is: [Abnormal noise from the engine intake valve, fault location: intake valve of cylinder 3, fault level: level 2, it is recommended to have the valve clearance checked and adjusted at an authorized Changan Automobile service station within 7 days].

[0032] Example 2:

[0033] Data Acquisition: Vibration Sensors: Three vibration sensors are fixed at the connection points between the engine's left and right shock absorbers and the vehicle's subframe, as well as at the connection point between the transmission and the engine shock absorber, to collect vibration transmission characteristic data; Microphone Array: One microphone is placed in the center of the bottom of the engine compartment, and another microphone is placed inside the left front wheel arch to collect abnormal noise data.

[0034] Actual vehicle conditions: The vehicle was in low-speed driving condition, with a speed of 30-40 km / h, unloaded, driving straight on a flat road, and the engine speed was 1500-2000 r / min. Vibration data: A vibration frequency of 120±10Hz was detected at the left engine damper pad. The vibration started synchronously with the engine speed increasing to 1500 r / min, with a vibration amplitude of 0.52 mm / s. The vibration was transmitted to the subframe of the vehicle body, resulting in obvious resonance. The vibration frequencies of the right damper pad and the transmission damper pad were normal, with amplitudes of <0.1 mm / s. Sound data: A muffled "humming" sound was collected in the 110-130Hz audio range. The tone was low, and a wide characteristic peak appeared at 120Hz in the spectrum. The timbre was a rubber friction + metal resonance sound. The sound persisted in the engine speed range of 1500-2000 r / min. The abnormal noise disappeared when the engine speed was below 1500 r / min or above 2000 r / min.

[0035] Fault matching and location: By comparing real-time data with the fault spectrum database of abnormal noises of Changan Oushang series passenger vehicles, the vibration frequency of about 120Hz and the dull "humming" sound of rubber friction are typical characteristics of the aging and elasticity loss of the engine damping pad rubber. Combined with the vibration sensor collection location (left engine damping pad) and vibration resonance transmission characteristics, the fault location is accurately located as the aging of the rubber layer of the hydraulic damping pad on the left side of the engine and the internal hydraulic oil leakage.

[0036] Fault Level Assessment: According to the fault level classification standard, this fault is a Level 1 fault (minor fault), which only causes slight resonance and abnormal noise in the vehicle body, poses no driving safety hazard, and does not affect the normal operation of the engine.

[0037] Warning output: Send a fault warning to the vehicle's infotainment system and the user's mobile app. The warning information is: [Abnormal noise from the engine damping pad, fault location: left hydraulic damping pad, fault level: Level 1, it is recommended to have the damping pad checked and replaced at the next routine vehicle maintenance (approximately 8000km remaining).]

[0038] Example 3:

[0039] Data acquisition: Two vibration sensors are fixed to the connection ends of the left and right front steering ball joints and tie rods, respectively, to collect data such as vibration frequency and vibration triggering timing; Microphone array: One waterproof and dustproof microphone is arranged on the inner side of each of the left and right front wheel arches to collect abnormal noise data under steering conditions.

[0040] Actual vehicle conditions: The vehicle was in a low-speed turning condition, traveling at 10-15 km / h, unloaded, making a left turn on a flat concrete road, with the steering wheel angle at 30°. Vibration data: An intermittent vibration frequency of 85±8Hz was detected at the left front steering ball joint. The vibration started synchronously with the steering wheel turning 30° to the left. The vibration occurred intermittently during the turn, with a single vibration duration of 0.5-1s and an amplitude of 0.48mm / s. No abnormal vibration data was found at the right front steering ball joint. Sound data: An intermittent "click" sound in the 80-90Hz audio range was collected at the left front wheel arch. The sound was crisp, with an intermittent spike at 85Hz in the spectrum. The timbre was a loose impact sound at the metal hinge. The sound only appeared when turning left, and there was no abnormal noise when driving straight or turning right.

[0041] Fault matching and location: By comparing real-time data with the fault spectrum database of abnormal noises of Changan Oushang series passenger vehicles, the intermittent vibration frequency of about 85Hz and the crisp "click" sound of metal hinge are typical characteristics of loose steering ball joint and excessive clearance between ball joint pin and ball cup. Combined with the vibration sensor acquisition location (left front steering ball joint), sound source (inside left front wheel arch), and abnormal noise triggering condition (left turn), the fault location was accurately located as loose left front steering tie rod outer ball joint, and the ball joint dust cover was damaged, resulting in grease loss.

[0042] Fault Level Assessment: According to the fault level classification standard, this fault is a Level 2 fault (low-level safety fault). Loose steering ball joints will affect the vehicle's steering precision. If it is used for a long time, it will cause the ball joint to wear out completely, leading to the risk of loss of steering control. It needs to be repaired as soon as possible.

[0043] Warning output: Send a fault warning to the vehicle-mounted infotainment system and the user's mobile app. The warning information is: [Steering system abnormal noise, fault location: left front steering tie rod outer ball joint, fault level: level 2, it is recommended to have the ball joint checked and replaced at a Changan Automobile authorized service station within 3 days, and perform four-wheel alignment calibration after replacement].

[0044] The above three cases were all diagnosed using the fault prediction and early warning method of the present invention, and were completely consistent with the actual vehicle disassembly and testing results of the authorized service station of Changan Automobile. This verifies the technical feasibility of the method of the present invention in diagnosing mechanical abnormal noise faults in automobiles, which is accurate in positioning, reliable in data matching, and reasonable in fault level determination. At the same time, relying on the passenger car general parts abnormal noise database established by Changan Automobile, the predictive judgment of faults is realized, which meets the industry development needs of predictive maintenance in automobiles.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for predicting and warning of automotive mechanical faults, characterized in that, The process includes the following steps: S1, acquiring vibration and sound data of vulnerable vehicle components under different vehicle conditions, and establishing a historical database of vulnerable vehicle components under different vehicle conditions; S2, importing the historical database of vulnerable components into a robot learning model, and determining the vehicle status based on the vibration and sound data of each vulnerable component under different vehicle conditions, wherein the vehicle status includes normal and abnormal states; S3, extracting the vibration and sound characteristics of each vulnerable component under abnormal vehicle conditions, under different vehicle conditions and at different fault stages, and drawing abnormal noise fault maps of different vulnerable components under different vehicle conditions based on the extracted feature data; S4, acquiring real-time vibration and sound data of each vulnerable component of the vehicle, matching the acquired real-time vibration and sound data with the abnormal noise fault maps, and determining the cause and location of the vehicle fault; S5, determining the fault level based on the cause and location of the vehicle fault, and issuing a fault warning to the vehicle or user terminal.

2. The method for predicting and warning of automotive mechanical faults according to claim 1, characterized in that, The vibration data is obtained through vibration sensors installed on each vulnerable component of the vehicle, and the sound data is obtained using microphone arrays installed on each vulnerable component; wherein, the vibration data includes vibration frequency, vibration duration and vibration start time, and the sound data includes audio, phase, pitch, timbre, spectrum and duration.

3. The method for predicting and warning of automotive mechanical faults according to claim 1 or 2, characterized in that, The vibration and sound data of the vehicle's vulnerable components under different vehicle conditions are vibration and sound data obtained from the vehicle's vulnerable components at different driving speeds, loads, road conditions, and driving states. The road conditions include flat roads, downhill sections, and uphill sections, and the driving states include going straight, turning left, and turning right.

4. The method for predicting and warning of automotive mechanical faults according to claim 3, characterized in that, The abnormal noise fault spectrum includes the audio, phase, pitch, timbre, spectrum, and duration of abnormal noises corresponding to the cause of the fault in different operating conditions when the corresponding fault occurs in different damaged components of the vehicle, as well as the real-time vibration frequency, vibration duration, and vibration start time corresponding to the cause of the fault in that operating condition.

5. The method for predicting and warning of automotive mechanical faults according to claim 4, characterized in that, The fault level is determined based on the abnormal noise vibration frequency, vibration duration, abnormal noise pitch, and abnormal noise duration, including Level 1 fault, Level 2 fault, and Level 3 fault. Level 1 fault is a minor fault that will not affect driving safety. Level 2 fault is a low-level safety fault that needs to be repaired as soon as possible. Level 3 fault is a high-level fault that needs to be repaired immediately.

6. The method for predicting and warning of automotive mechanical faults according to claim 5, characterized in that, The location of the abnormal noise fault is determined by the vehicle condition, sound phase, pitch, and sound source at the time the abnormal noise occurs.

Images