Engine fault monitoring system and method based on sound signals

Abstract

The invention relates to the technical field of engine monitoring, in particular to an engine fault monitoring system and method based on sound signals. Comprising an equipment supporting and protecting device, a jacking device, a sound collecting device, a rotating speed sensor, a silencer device, a data transmission module and a data analysis module. Through synchronous acquisition of high-precision sound and rotating speed signals, automatic jacking positioning and noise reduction hardware cooperation and an intelligent data analysis module, the accuracy and real-time performance of engine fault diagnosis are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of engine monitoring technology, and in particular to an engine fault monitoring system and method based on sound signals. Background Technology

[0002] As the core power component of various mechanical equipment and vehicles, the engine's operating status directly affects the reliability, safety, and service life of the entire system. Therefore, engine fault monitoring and diagnosis technology has always been a key research focus in the field of industrial testing and maintenance.

[0003] In recent years, with the advancement of acoustic signal processing and pattern recognition technologies, sound signal-based fault diagnosis has emerged as a promising non-invasive monitoring method. The sound waves generated by an engine during operation contain rich state information, reflecting various physical processes such as internal mechanical friction, gas dynamics, and combustion. Compared to vibration signals, sound signal acquisition is more convenient, sensor (microphone) placement is flexible, and it can simultaneously perceive the combined effects of multiple potential fault sources. Therefore, it exhibits unique advantages in comprehensive engine condition assessment and early fault warning.

[0004] However, applying sound signal analysis technology to engine fault monitoring, especially to achieve high-precision and high-reliability online diagnosis, still faces a series of key technical challenges: In actual industrial or testing environments, engine sound signals are easily drowned out by strong exhaust noise, background environmental noise, and other equipment operating noise, resulting in a low signal-to-noise ratio of the collected signals and making it difficult to extract effective fault features.

[0005] The acoustic characteristics of an engine are closely related to its operating conditions, such as engine speed and load. Many existing methods do not precisely synchronize sound signals with engine speed signals when collecting them, resulting in analysis results that cannot accurately correspond to specific operating conditions, thus reducing the specificity and accuracy of diagnostics.

[0006] Many existing methods still rely on technicians' visual observation and experience-based judgment of acoustic spectra or characteristic parameters, lacking automated feature extraction, fault mode identification, and decision-making mechanisms. They are inefficient and inconsistent, making it difficult to meet the needs of real-time online monitoring.

[0007] Existing sound monitoring solutions are often designed for specific models or laboratory environments, with complex hardware deployments and a lack of auxiliary devices for rapid and accurate engine positioning and noise isolation. This results in inconvenient system installation, poor versatility, and difficulty in promoting and applying the solutions in diverse scenarios.

[0008] In addition, although there are existing technologies that use electrical signal analysis to determine the probability of failure (for example, see patent document CN104807642A), they usually rely on professional electrical signal sensors, require intrusive access to the engine circuit, have blind spots in monitoring certain mechanical faults, and have complex overall system processes and high costs.

[0009] In summary, existing engine fault monitoring technologies, especially those based on sound signal analysis, still have significant shortcomings in terms of noise interference resistance, signal acquisition accuracy and synchronization, automation of the diagnostic process, and system practicality and adaptability. Summary of the Invention

[0010] This invention provides an engine fault monitoring system and method based on sound signals. See the description below for details: An engine fault monitoring system based on sound signals includes: an equipment support and protection device, a lifting device, a sound acquisition device, a speed sensor, a muffler device, a data transmission module, and a data analysis module; The equipment support and protection device constitutes the main support structure and protective cavity of the system; The lifting device is installed in the bottom working area of ​​the equipment support and protection device, and is used to carry and lift the engine to the predetermined working position; The sound acquisition device is installed above the engine's workstation and is used to collect sound signals when the engine is running. The speed sensor is set at the speed detection part of the engine and is used to synchronously collect the engine speed signal; The muffler device is arranged around the engine's workstation to reduce the interference of environmental noise on sound signal acquisition. The data transmission module is used to transmit the collected sound signals and rotation speed signals to the data analysis module; The data analysis module is used to process and analyze the received sound signals and speed signals in order to diagnose engine faults.

[0011] Preferably, the lifting device is provided with a limiting block for defining the engine placement position and a laser sensor for detecting the engine's positioning status. The lifting device performs a lifting action based on the detection signal from the laser sensor.

[0012] Preferably, it further includes a positioning control device, which is installed in the area above the inner workstation of the equipment support and protection device. The sound acquisition device is fixedly installed on the execution end of the positioning control device, and the speed sensor is installed on the corresponding execution component of the positioning control device.

[0013] Preferably, the sound acquisition device includes one or more microphones arranged at measurement points at the cylinder head and / or housing of the engine.

[0014] Preferably, the muffler device is connected to the engine's exhaust port and automatically connects after the engine reaches the operating position.

[0015] Preferably, the data transmission module is a multi-channel high-speed data acquisition device.

[0016] Preferably, it also includes a status display interface for displaying the device's operating status and parameters and / or a tri-color light for indicating the device's operating status, wherein the status display interface and the tri-color light are mounted on the device support and protection device.

[0017] The present invention also discloses an engine fault monitoring method using the system, comprising the following steps: S1: Place the engine on the lifting device, detect and position it using the laser sensor 7, and then lift the engine to the working position using the lifting device; S2: The muffler automatically connects to the engine exhaust port to create a noise-reducing environment; S3: Collects the sound signal of the engine running through the sound acquisition device, and simultaneously collects the engine speed signal through the speed sensor; S4: The collected sound signal and rotation speed signal are sent to the data analysis module in real time through the data transmission module; S5: The data analysis module performs synchronous analysis and processing of sound signals and speed signals, diagnoses engine status based on preset fault feature models, and outputs fault diagnosis results.

[0018] Preferably, in step S5, the data analysis module uses one or more of the following methods—time domain analysis, frequency domain analysis, and envelope analysis—to process the sound signal and combines it with the rotational speed signal to perform operating condition correlation analysis.

[0019] Preferably, the fault diagnosis results output in step S5 are displayed and alarmed in real time through a status display interface and / or a three-color light.

[0020] The beneficial effects of the technical solution provided by this invention are as follows: This invention significantly improves the accuracy and real-time performance of engine fault diagnosis through high-precision synchronous acquisition of sound and speed signals, automated lifting and positioning, noise reduction hardware coordination, and an intelligent data analysis module. It effectively overcomes industry challenges such as strong environmental noise interference, weak correlation between signals and operating conditions, and reliance on human experience, achieving non-intrusive, automated, and highly reliable condition monitoring and early warning. The system possesses good adaptability and scalability, requires no modification to the engine structure, is easy to deploy, and significantly reduces monitoring costs and maintenance downtime, providing an efficient and practical technical solution for predictive engine maintenance. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall structure of the engine fault monitoring system based on sound signals provided by the present invention; Figure 2 This is a schematic diagram showing the connection and arrangement of the sound acquisition device and the speed signal acquisition device in this invention; Figure 3 This is a schematic diagram of the core signal acquisition device involved in this invention; Figure 4 This is a schematic diagram of the structure and operation of the lifting device in this invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0023] The present invention provides an engine fault monitoring system based on sound signals. Its core lies in achieving the synchronous acquisition of high-quality sound signals and accurate speed signals through a highly integrated and automated hardware platform, and using intelligent algorithms for real-time analysis and diagnosis.

[0024] The main hardware of the system is built on a rigid equipment support and protection device 3. This device is usually constructed by welding or bolting steel to form a frame, and the outer cover plate forms a closed or semi-closed cavity, which has the functions of support, protection and preliminary sound insulation.

[0025] At the testing station area at the bottom of the supporting protective device 3, a lifting device 5 is fixedly installed with bolts. This lifting device 5 preferably employs a pneumatic or electric screw lifting mechanism. Limiting blocks 6 for physical positioning are welded or bolted to its supporting platform, and at least one laser sensor 7 is installed at the edge of the platform. During implementation, the operator places the engine 4 to be tested on the lifting platform, and the laser sensor 7 detects the engine's outline or specific markings. When the first sensor detects the engine approaching, the system pre-determines its position; after the second sensor confirms that the engine has completely entered the preset area, the lifting device 5 is activated, smoothly lifting the engine to a preset height (e.g., aligning the engine flywheel or output shaft centerline with the sound acquisition device), and holding it locked to ensure absolute stability of the engine position during the testing process.

[0026] Above the support and protection device 3, a positioning control device 9 is mounted via a bracket. This device can be a multi-degree-of-freedom robotic arm or a precision linear module. A sound acquisition device 8 (whose core is one or more high-precision microphones 10, such as PCB378B02) is rigidly mounted at the end of the positioning control device 9. During implementation, the positioning control device 9 is driven by a control program to precisely move the microphones 10 to predetermined measuring points above the engine 4, such as directly opposite the center of each cylinder head or the side of the crankcase, approximately 10-50mm from the engine surface. Simultaneously, a speed sensor 12 (such as a photoelectric or Hall effect sensor) is also mounted on the same actuator or on a nearby independent bracket, with its probe precisely aligned with the engine flywheel gear, ignition high-voltage wire, or magneto signal disk—locations that reflect real-time speed. This ensures that the sound and speed signals maintain spatial and temporal correlation at the physical acquisition source.

[0027] A muffler device 11 is installed around the engine bay on the inner wall of the supporting protective device 3. This device can be designed as a movable baffle with sound-absorbing material or a retractable muffler cover. In practice, when the engine is lifted into position, the drive mechanism (such as a cylinder) automatically closes the muffler baffle or pushes the muffler cover to the engine exhaust port and achieves a sealed connection, thereby creating a localized high-quiet environment around the engine and significantly reducing exhaust noise and reflected noise.

[0028] The output signal cables of the sound acquisition device 8 and the speed sensor 12 are connected to a high-speed data acquisition card (such as NI-4431), which forms the core of the data transmission module. The acquisition card is installed in a slot of the industrial control computer or connected to it via a dedicated interface. The key to its implementation lies in configuring a unified sampling clock to ensure strict synchronization of data sampling across all channels (sound channel and speed channel). The sampling rate is set according to the engine's highest operating frequency, typically not lower than 40kHz, to meet the requirements for capturing high-frequency components of fault characteristics.

[0029] The data analysis module runs as software on the system's industrial control computer. Its implementation process is as follows: a) Signal Synchronization and Preprocessing: The software receives the synchronization data stream from the acquisition card. First, the speed signal is decoded to obtain a precise instantaneous speed sequence and cycle markers (such as pulses per revolution). Using these sequence markers, the audio signal is cyclically segmented and aligned to achieve synchronous analysis according to the working cycle (or crankshaft angle), eliminating the influence of speed fluctuations.

[0030] b) Feature extraction: For each audio signal segment within a loop, perform multiple analyses in parallel: Time-domain analysis: Calculates RMS value, peak value, kurtosis, impulse factor, etc., to reflect the overall energy and impulse components of the signal.

[0031] Frequency domain analysis: Perform Fast Fourier Transform (FFT) to obtain the spectrum. Focus on the frequency band energy related to fault characteristics such as engine combustion frequency, piston slap frequency, and bearing passage frequency.

[0032] Time-frequency domain analysis: For non-stationary signal segments, short-time Fourier transform (STFT) or wavelet transform is used to observe the change of characteristic frequencies with time (or crankshaft angle).

[0033] Envelope analysis: The sound signal is bandpass filtered (around the characteristic frequencies of bearing or gear faults), Hilbert transform is used to extract the envelope, and then the envelope is subjected to spectral analysis to diagnose early minor impact faults.

[0034] c) Fault Diagnosis and Decision Making: The extracted multi-dimensional feature vectors are input into a pre-trained fault diagnosis model. This model can be a rule base based on threshold judgment or a machine learning model (such as Support Vector Machine (SVM) or deep learning network). The model integrates sound features and their correlation patterns with engine speed to output an assessment of the current engine health status, possible fault types (such as valve leakage, piston pin loosening, bearing wear, etc.), and confidence levels.

[0035] d) Result Feedback: The diagnostic results are displayed in real time on the status display interface 2, including speed, characteristic curve, fault alarm information, maintenance suggestions, etc. At the same time, the three-color indicator 1 is driven to change color (e.g., green for normal, yellow for warning, and red for alarm), providing remote status indication.

[0036] Taking the testing of a single-cylinder engine as an example: Step 1: The operator pushes the engine into the equipment support and protection device 3, placing it on the platform of the lifting device 5. The laser sensor 7 detects the engine's position, the lifting device 5 is activated, raising the engine and locking it at the detection height.

[0037] Step 2: The positioning control device 9 automatically moves the probe, which integrates the microphone 10 and the speed sensor 12, to a predetermined position directly above the cylinder head. The muffler device 11 automatically closes, covering the exhaust port.

[0038] Step 3: Start the engine and stabilize it at a specific operating condition (such as rated speed). The system simultaneously collects sound and speed signals.

[0039] Step 4: The acquisition card synchronously acquires the two signals at a rate of 102.4 kS / s and transmits them to the industrial control computer in real time.

[0040] Step 5: The data analysis software synchronizes, segments, and extracts kurtosis indices and energy values ​​for specific frequency bands (e.g., 2-5kHz). It is found that the kurtosis value consistently exceeds the threshold and the energy in that frequency band increases significantly, consistent with the fault characteristic library pattern of "loose piston pin."

[0041] Step Six: The system displays a "Piston pin suspected of being loose" alarm on status display interface 2 and switches the tri-color indicator light 1 to red. Simultaneously, it records complete raw data and diagnostic logs.

[0042] The present invention is not limited to the specific embodiments described above. For example, the number of microphones 10 in the sound acquisition device 8 can be flexibly increased or decreased according to the number of engine cylinders; the muffler device 11 can adopt active noise reduction technology; the fault diagnosis model of the data analysis module can be updated and optimized regularly through cloud data; the system can also be extended to connect to the factory's MES (Manufacturing Execution System) to realize the automatic dispatch of maintenance work orders.

[0043] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An engine fault monitoring system based on sound signals, characterized in that, include: Equipment support and protection device (3), lifting device (5), sound acquisition device (8), speed sensor (12), silencer device (11), data transmission module and data analysis module; The equipment support and protection device (3) constitutes the main support structure and protective cavity of the system; The lifting device (5) is installed in the bottom working area of ​​the equipment support and protection device (3) to carry and lift the engine (4) to the predetermined working position; The sound acquisition device (8) is set above the work position of the engine (4) and is used to acquire the sound signal when the engine is running. The speed sensor (12) is set at the speed detection part of the engine (4) and is used to synchronously collect the speed signal of the engine; The muffler device (11) is arranged around the work station of the engine (4) to reduce the interference of environmental noise on the sound signal acquisition; The data transmission module is used to transmit the collected sound signals and rotation speed signals to the data analysis module; The data analysis module is used to process and analyze the received sound signals and speed signals in order to diagnose engine faults.

2. The engine fault monitoring system according to claim 1, characterized in that, The lifting device (5) is provided with a limiting block (6) for limiting the engine placement position and a laser sensor (7) for detecting the engine's position. The lifting device (5) performs a lifting action according to the detection signal of the laser sensor (7).

3. The engine fault monitoring system according to claim 1, characterized in that, It also includes a positioning control device (9), which is installed in the area above the inner workstation of the equipment support and protection device (3). The sound acquisition device (8) is fixedly installed on the execution end of the positioning control device (9), and the speed sensor (12) is installed on the corresponding execution component of the positioning control device (9).

4. The engine fault monitoring system according to claim 1, characterized in that, The sound acquisition device (8) includes one or more microphones (10) arranged at measurement points at the cylinder head and / or housing of the engine.

5. The engine fault monitoring system according to claim 1, characterized in that, The muffler device (11) is connected to the exhaust port of the engine and automatically connects after the engine reaches the working position.

6. The engine fault monitoring system according to claim 1, characterized in that, The data transmission module is a multi-channel high-speed data acquisition device.

7. The engine fault monitoring system according to any one of claims 1 to 6, characterized in that, It also includes a status display interface (2) for displaying the operating status and parameters of the equipment and / or a tri-color light (1) for indicating the operating status of the equipment, wherein the status display interface (2) and the tri-color light (1) are installed on the equipment support and protection device (3).

8. An engine fault monitoring method using the system described in any one of claims 1 to 7, characterized in that, Includes the following steps: S1: Place the engine (4) on the lifting device (5), detect and position it by the laser sensor (7), and lift the engine to the working position by the lifting device (5); S2: The muffler device (11) is automatically connected to the engine exhaust port to create a noise reduction environment; S3: The sound signal of the engine operation is collected by the sound acquisition device (8), and the speed signal of the engine is collected synchronously by the speed sensor (12); S4: The collected sound signal and rotation speed signal are sent to the data analysis module in real time through the data transmission module; S5: The data analysis module performs synchronous analysis and processing of sound signals and speed signals, diagnoses engine status based on preset fault feature models, and outputs fault diagnosis results.

9. The engine fault monitoring method according to claim 8, characterized in that, In step S5, the data analysis module uses one or more of the following methods—time domain analysis, frequency domain analysis, and envelope analysis—to process the sound signal and combines it with the speed signal to perform operating condition correlation analysis.

10. The engine fault monitoring method according to claim 8, characterized in that, The fault diagnosis results output in step S5 are displayed and alarmed in real time through the status display interface (2) and / or the three-color light (1).

Images