An internal combustion engine combustion noise identification technology based on EFD-SST

Abstract

The application discloses an internal combustion engine combustion noise identification technology based on EFD-SST, and belongs to the technical field of internal combustion engines. The internal combustion engine combustion noise identification technology comprises the following steps: step S1: respectively testing the whole machine 5-point noise signals of the internal combustion engine in the rated condition and the motoring condition, wherein the rated condition is 2000r / min, and the motoring condition is 2000r / min; and step S2: separating mechanical noise from combustion noise, and separating the noise by using the motoring method. In the application, the separation of mechanical noise from combustion noise is completed by testing the internal combustion engine noise data and using the motoring method, the combustion noise signal is decomposed by using an empirical Fourier decomposition (EFD) and a synchronous wavelet compression transformation (SST), the internal mechanism of the combustion noise is analyzed, the precision and efficiency of the internal combustion engine combustion noise identification are improved, and more accurate and effective guidance is provided for the control of the internal combustion engine combustion noise.

Description

Technical Field

[0001] This invention belongs to the field of internal combustion engine technology, and particularly relates to an internal combustion engine combustion noise identification technology based on EFD-SST. Background Technology

[0002] Internal combustion engine noise is categorized into mechanical noise, aerodynamic noise, and combustion noise. Combustion noise plays a significant role in the overall noise level of an internal combustion engine; therefore, with advancements in internal combustion engine technology, research on combustion noise has become increasingly in-depth. Combustion noise is generally defined as the noise generated by the dynamic load and high-frequency vibrations caused by the rapid increase in cylinder pressure during fuel combustion, transmitted through the cylinder head, cylinder liner, piston-connecting rod-crankshaft, and main bearings to the engine block, as well as the noise radiated from the engine's structural surfaces due to vibrations caused by the cylinder head.

[0003] During normal combustion in a gasoline engine, combustion noise accounts for a relatively small proportion of the total noise. However, the noise generated during abnormal combustion, such as knocking and surface ignition, must be taken seriously. During knocking, the gas pressure inside the cylinder rises sharply, generating high-frequency knocking noise of 4000Hz–6000Hz. In high-compression gasoline engines, excessive carbon deposits can lead to overheating and surface ignition, resulting in a surge in cylinder pressure and generating noise of 500Hz–2000Hz. However, for gasoline engines in normal production and use, knocking noise can be avoided by selecting the appropriate gasoline grade and ignition advance angle; surface ignition can be eliminated simply by cleaning carbon deposits in the combustion chamber. Therefore, identifying the internal mechanisms of combustion noise in internal combustion engines provides comprehensive technical support for understanding the characteristics of combustion noise and reducing it.

[0004] Since both mechanical noise and combustion noise originate inside the internal combustion engine and radiate outwards through its structural surface, they are intertwined in terms of timing, location, propagation path, and basic frequency domain characteristics. Therefore, identifying and separating these two types of noise is crucial. Currently, the most commonly used methods in China are separate operation and blind source separation techniques. However, blind source separation techniques often only produce waveforms similar to the original signal, with amplitude information often being meaningless.

[0005] Regarding signal decomposition research, some scholars have proposed Empirical Mode Decomposition (EMD), bringing time-frequency analysis to a new level. Essentially, EMD algorithms stationarize signals, decomposing them into Intrinsic Mode Functions (IMFs) for easy analysis. However, EMD suffers from mode aliasing and endpoint effects during signal decomposition, significantly interfering with the separation of noise features. Other scholars have proposed Variational Mode Decomposition (VMD), an algorithm based on signal theory that adaptively partitions the signal according to the frequency domain. Compared to traditional EMD methods, VMD exhibits better noise robustness and effectively avoids mode aliasing. However, VMD requires pre-setting parameters before signal processing; inaccurate parameter selection can introduce interference signals during decomposition, leading to errors in feature extraction.

[0006] Currently, commonly used linear time-frequency analysis methods in internal combustion engine signal analysis include Short Time Fourier Transform (STFT) and Continuous Wavelet Transform (CWT). STFT, due to its fixed window function, consistently fails to resolve the conflict between time and frequency resolution. CWT employs a flexible window function with a fixed window size but varying shape; the window length is smaller at high frequencies, providing higher time-domain resolution, while the window length is larger at low frequencies, providing higher frequency resolution. However, its time-frequency resolution is limited by Heisenberg uncertainty, preventing it from achieving optimal resolution simultaneously.

[0007] In existing technologies, the internal mechanisms of combustion noise generation include drastic pressure changes within the cylinder, high-frequency gas vibrations, and gas pressure vibration excitation. Currently, signal processing techniques are used for internal combustion engine combustion noise identification, but these methods suffer from low identification efficiency and poor separation performance. To address these issues, this paper proposes an internal combustion engine combustion noise identification technology based on EFD-SST to improve identification accuracy and efficiency, providing more precise and effective guidance for the control of internal combustion engine combustion noise. Summary of the Invention

[0008] The purpose of this invention is to address the problems in the prior art where the internal mechanisms of combustion noise generation include drastic pressure changes within the cylinder, high-frequency gas vibration, and gas pressure vibration excitation. Furthermore, to address the issues of low identification efficiency and poor separation effect when using signal processing technology for internal combustion engine noise identification, this invention proposes an internal combustion engine noise identification technology based on EFD-SST.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] An internal combustion engine combustion noise identification technology based on EFD-SST, the internal combustion engine combustion noise identification technology includes the following steps:

[0011] S1: Test the noise signals of the internal combustion engine at 5 points under standard conditions and reverse towing conditions respectively. The standard condition is 2000 r / min and the reverse towing condition is 2000 r / min.

[0012] S2: The total noise and mechanical noise of the internal combustion engine are separated by using the reverse drag method. One-third harmonic processing is performed to obtain the noise value corresponding to each center frequency. The combustion noise spectrum of the internal combustion engine is obtained by using the total noise value and mechanical noise value at the center frequency.

[0013] S3: Based on EFD technology, the combustion noise signal of the internal combustion engine is decomposed;

[0014] S4: Based on SST, the intrinsic mode function (IMF) of the noise decomposition components is obtained using IMF time-frequency analysis technology.

[0015] As a further description of the above technical solution:

[0016] The sampling frequency of the noise measurement point is 20480Hz. The noise test is conducted in a semi-anechoic chamber with dimensions of 11.2m×9.5m×6.0m, a cutoff frequency of 100Hz, and a background noise of 25dB.

[0017] As a further description of the above technical solution:

[0018] The noise measurement equipment includes the AVL internal combustion engine test system, the LMS Test.lab 32-bit acoustic vibration test system, the BSWA acoustic microphone, and the PCB acceleration sensor.

[0019] As a further description of the above technical solution:

[0020] The separation of mechanical noise and combustion noise specifically includes:

[0021] In the method of noise separation using the reverse dragging method, the overall noise includes combustion noise and mechanical noise. The experimentally measured reverse dragging noise is regarded as mechanical noise. The experimentally measured overall noise and reverse dragging noise are used to calculate the internal combustion engine combustion noise using the sound level superposition method.

[0022] For n independent sound sources existing simultaneously, according to the superposition of energy and the relationship between sound power W and sound pressure P: W∝p 2 The total sound pressure level p can be obtained. t The calculation formula is:

[0023]

[0024] In the formula, p i (i = 1, 2, ..., n) represents the sound pressure level of each sound source;

[0025] According to the definition of sound pressure level, the expression for total sound pressure level is:

[0026]

[0027] In the formula, p0 is the reference sound pressure level, which is usually expressed as p0 = 2 × 10⁻⁶. -5 MPa;

[0028] From the above two equations, we can obtain:

[0029]

[0030] Total sound pressure level L pt With the sound pressure level L of each sound source pi The relationship between them can be represented as:

[0031]

[0032] Let the two sound pressure levels be L1 and L2. If the total sound pressure level L and one of the sound pressure levels L1 are known, then the other sound pressure level L2 can be calculated:

[0033]

[0034] By performing one-third harmonic processing on the total noise and mechanical noise of the internal combustion engine, the noise value corresponding to each center frequency can be obtained. The sound level is then decomposed using the total noise value and mechanical noise value at that center frequency to obtain the combustion noise spectrum of the internal combustion engine.

[0035] As a further description of the above technical solution:

[0036] The Fourier spectrum of the signal f(t) to be decomposed is obtained by Fourier transform.

[0037]

[0038] Using improved Fourier spectrum segmentation techniques to The sequence is divided and sorted, and the frequencies corresponding to the first N maximum values ​​in the sorted sequence are [Ω1, Ω2, ... Ω]. N The boundaries of the segments are defined by ω. n Sure:

[0039]

[0040] In the formula, Represents Ω n To Ω n+1 The Fourier spectrum amplitudes between;

[0041] Based on the obtained ω n Establish a zero-phase filter bank

[0042]

[0043] use Obtain the filtered signal

[0044]

[0045] Through calculation in step four The inverse Fourier transform yields different intrinsic mode functions (IMFs) and f. n (t):

[0046]

[0047] As a further description of the above technical solution:

[0048] The generation of combustion noise in internal combustion engines can be attributed to three main mechanistic factors:

[0049] The sudden pressure change inside the cylinder causes the cylinder and all its connected components to be subjected to strong impact dynamic loads, forming instantaneous excitation, which is transmitted to the entire internal combustion engine through the cylinder head, cylinder liner, piston connecting rod mechanism, crankshaft and its support, resulting in structural vibration of each component.

[0050] During the ignition delay period, the fuel injected into the combustion chamber ignites at several points in the combustion chamber when the combustion conditions are ripe. As a result, the local pressure at the initial ignition point rises sharply, and the flame spreads to the adjacent area. At the same time as the flame spreads, pressure waves with impact characteristics also spread. When these pressure waves encounter the combustion chamber wall, they are reflected. The repeated reflections of the pressure waves in the cylinder generate high-frequency vibrations in the gas. These high-frequency vibrations are maintained for a considerable period of time during the expansion stroke, thereby radiating high-frequency noise.

[0051] The gas pressure inside the cylinder changes periodically under stable operating conditions. By performing a Fourier transform on the cylinder pressure curve, a frequency domain pressure amplitude spectrum with the number of internal combustion engine working cycles per second as the fundamental frequency can be obtained. This constitutes the gas pressure vibration excitation, which will excite the forced vibration of the components connected inside the internal combustion engine cylinder.

[0052] As a further description of the above technical solution:

[0053] The SST-based IMF time-frequency analysis specifically includes:

[0054] The time-frequency characteristics of the IMF are obtained by using the synchronous wavelet compression transform (SST). The synchronous wavelet compression transform is based on the CWT and compresses and rearranges the complex coefficient spectrum of the CWT along the scale direction, which improves the resolution of the CWT while maintaining its complete mathematical invertibility.

[0055] Based on the time-frequency analysis results, the main excitation of internal combustion engine combustion noise can be preliminarily determined;

[0056] Based on prior knowledge of internal combustion engines, the coherence of cylinder pressure and noise, and the results of the reverse drag test, the separation of combustion noise is further determined.

[0057] As a further description of the above technical solution:

[0058] The system is capable of effectively identifying the main excitations of internal combustion engine combustion noise, including drastic changes in cylinder pressure, high-frequency vibrations of gas, and gas pressure vibration excitations.

[0059] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0060] 1. In this invention, by testing internal combustion engine noise data, the mechanical noise and combustion noise are separated using the reverse drag method. The combustion noise signal is decomposed using Empirical Fourier Decomposition (EFD)-Synchronous Wavelet Compressed Transform (SST), and the internal mechanism of combustion noise is analyzed. This results in high identification efficiency and good separation effect during the identification of internal combustion engine combustion noise, improving the accuracy and efficiency of identification and providing more accurate and effective guidance for the control of internal combustion engine combustion noise.

[0061] 2. In this invention, EFD combines improved Fourier spectrum segmentation technology and zero-phase filter bank. Compared with empirical mode decomposition and variational mode decomposition methods, the EFD method provides accurate and consistent decomposition results for multiple non-stationary modes and near-range mode signals, and can obtain accurate time-frequency representation results. It also has the highest computational efficiency. The phase-based synchronous compressed wavelet transform method SWT is used as the basis. The complex coefficient spectrum of CWT is compressed and rearranged along the scale direction, which improves the resolution and maintains complete mathematical invertibility. It is widely used in the engineering field for dynamic signal extraction and instantaneous frequency extraction of structures. Attached Figure Description

[0062] Figure 1 This is a flowchart of an internal combustion engine combustion noise identification technology based on EFD-SST proposed in this invention;

[0063] Figure 2 This is a schematic diagram of the location of the internal combustion engine noise test point in an internal combustion engine combustion noise identification technology based on EFD-SST proposed in this invention. Detailed Implementation

[0064] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] Please see Figure 1-2 This invention provides a technical solution: a method for identifying internal combustion engine combustion noise based on EFD-SST. The method utilizes EFD technology to adaptively decompose the combustion noise signal of an internal combustion engine and combines it with SST to identify each component. The method includes the following steps:

[0066] S1: Test the noise signals of the internal combustion engine at 5 points under standard conditions and reverse towing conditions respectively. The standard condition is 2000 r / min and the reverse towing condition is 2000 r / min.

[0067] S2: The total noise and mechanical noise of the internal combustion engine are separated by using the reverse drag method. One-third harmonic processing is performed to obtain the noise value corresponding to each center frequency. The combustion noise spectrum of the internal combustion engine is obtained by using the total noise value and mechanical noise value at the center frequency.

[0068] S3: Based on EFD technology, the combustion noise signal of the internal combustion engine is decomposed;

[0069] S4: Based on SST, the intrinsic mode function (IMF) of the noise decomposition components is obtained using IMF time-frequency analysis technology.

[0070] Specifically, the sampling frequency of the noise measurement points was 20480Hz, and the noise test was conducted in a semi-anechoic chamber with dimensions of 11.2m×9.5m×6.0m, a cutoff frequency of 100Hz, and a background noise of 25dB.

[0071] Specifically, the noise measurement equipment includes the AVL internal combustion engine test system, the LMS Test.lab 32-bit acoustic vibration test system, the BSWA acoustic microphone, and the PCB acceleration sensor.

[0072] Specifically, the separation of mechanical noise and combustion noise includes:

[0073] In the method of noise separation using the reverse dragging method, the overall noise includes combustion noise and mechanical noise. The experimentally measured reverse dragging noise is regarded as mechanical noise. The experimentally measured overall noise and reverse dragging noise are used to calculate the internal combustion engine combustion noise using the sound level superposition method.

[0074] For n independent sound sources existing simultaneously, according to the superposition of energy and the relationship between sound power W and sound pressure P: W∝p2 The formula for calculating the total sound pressure level (pt) can be derived as follows:

[0075]

[0076] In the formula, p i (i = 1, 2, ..., n) represents the sound pressure level of each sound source;

[0077] According to the definition of sound pressure level, the expression for total sound pressure level is:

[0078]

[0079] In the formula, p0 is the reference sound pressure level, which is usually expressed as p0 = 2 × 10⁻⁶. -5 MPa;

[0080] From the above two equations, we can obtain:

[0081]

[0082] Total sound pressure level L pt With the sound pressure level L of each sound source pi The relationship between them can be represented as:

[0083]

[0084] Let the two sound pressure levels be L1 and L2. If the total sound pressure level L and one of the sound pressure levels L1 are known, then the other sound pressure level L2 can be calculated:

[0085]

[0086] By performing one-third harmonic processing on the total noise and mechanical noise of the internal combustion engine, the noise value corresponding to each center frequency can be obtained. The sound level is then decomposed using the total noise value and mechanical noise value at that center frequency to obtain the combustion noise spectrum of the internal combustion engine.

[0087] Specifically, adaptive decomposition includes:

[0088] The Fourier spectrum of the signal f(t) to be decomposed is obtained by Fourier transform.

[0089]

[0090] Using improved Fourier spectrum segmentation techniques to The sequence is divided and sorted, and the frequencies corresponding to the first N maximum values ​​in the sorted sequence are [Ω1, Ω2, ... Ω]. N The boundaries of the segments are defined by ω. n Sure:

[0091]

[0092] In the formula, Represents Ω n To Ω n+1 The Fourier spectrum amplitudes between;

[0093] Based on the obtained ω n Establish a zero-phase filter bank

[0094]

[0095] use Obtain the filtered signal

[0096]

[0097] Through calculation in step four The inverse Fourier transform yields different intrinsic mode functions (IMFs) and f. n (t):

[0098]

[0099] Specifically, the generation of combustion noise in internal combustion engines can be attributed to three main mechanistic factors:

[0100] The sudden pressure change inside the cylinder causes the cylinder and all its connected components to be subjected to strong impact dynamic loads, forming instantaneous excitation, which is transmitted to the entire internal combustion engine through the cylinder head, cylinder liner, piston connecting rod mechanism, crankshaft and its support, resulting in structural vibration of each component.

[0101] During the ignition delay period, the fuel injected into the combustion chamber ignites at several points in the combustion chamber when the combustion conditions are ripe. As a result, the local pressure at the initial ignition point rises sharply, and the flame spreads to the adjacent area. At the same time as the flame spreads, pressure waves with impact characteristics also spread. When these pressure waves encounter the combustion chamber wall, they are reflected. The repeated reflections of the pressure waves in the cylinder generate high-frequency vibrations in the gas. These high-frequency vibrations are maintained for a considerable period of time during the expansion stroke, thereby radiating high-frequency noise.

[0102] The gas pressure inside the cylinder changes periodically under stable operating conditions. By performing a Fourier transform on the cylinder pressure curve, a frequency domain pressure amplitude spectrum with the number of internal combustion engine working cycles per second as the fundamental frequency can be obtained. This constitutes the gas pressure vibration excitation, which will excite the forced vibration of the components connected inside the internal combustion engine cylinder.

[0103] Specifically, the time-frequency analysis of the IMF based on SST is as follows:

[0104] The time-frequency characteristics of the IMF are obtained by using the synchronous wavelet compression transform (SST). The synchronous wavelet compression transform is based on the CWT and compresses and rearranges the complex coefficient spectrum of the CWT along the scale direction, which improves the resolution of the CWT while maintaining its complete mathematical invertibility.

[0105] Based on the time-frequency analysis results, the main excitation of internal combustion engine combustion noise can be preliminarily determined;

[0106] Based on prior knowledge of internal combustion engines, the coherence of cylinder pressure and noise, and the results of the reverse drag test, the separation of combustion noise is further determined.

[0107] Specifically, it can effectively identify the main excitations of internal combustion engine combustion noise, including drastic changes in cylinder pressure, high-frequency vibrations of gas, and gas pressure vibration excitations.

[0108] Working principle: When using,

[0109] S1: Test the noise signals of the internal combustion engine at 5 points under standard conditions and reverse towing conditions respectively. The standard condition is 2000 r / min and the reverse towing condition is 2000 r / min.

[0110] The sampling frequency of the noise measurement points was 20480Hz. The noise test was conducted in a semi-anechoic chamber with dimensions of 11.2m×9.5m×6.0m, a cutoff frequency of 100Hz, and a background noise of 25dB.

[0111] The main noise measurement equipment includes the AVL internal combustion engine test system, the LMS Test.lab 32-bit acoustic vibration test system, the BSWA acoustic microphone, and the PCB accelerometer.

[0112] S2: Separate mechanical noise and combustion noise. Use the reverse drag method to separate noise. By performing one-third harmonic processing on the total noise and mechanical noise of the internal combustion engine, the noise value corresponding to each center frequency can be obtained. Use the total noise value and mechanical noise value at the center frequency to decompose the sound level to obtain the combustion noise spectrum of the internal combustion engine.

[0113] The separation of mechanical noise and combustion noise specifically includes:

[0114] In the method of noise separation using the reverse dragging method, the overall noise includes combustion noise and mechanical noise. The experimentally measured reverse dragging noise is regarded as mechanical noise. The experimentally measured overall noise and reverse dragging noise are used to calculate the internal combustion engine combustion noise using the sound level superposition method.

[0115] For n independent sound sources existing simultaneously, according to the superposition of energy and the relationship between sound power W and sound pressure P: W∝p 2 The total sound pressure level p can be obtained. t The calculation formula is:

[0116]

[0117] In the formula, p i (i = 1, 2, ..., n) represents the sound pressure level of each sound source;

[0118] According to the definition of sound pressure level, the expression for total sound pressure level is:

[0119]

[0120] In the formula, p0 is the reference sound pressure level, which is usually expressed as p0 = 2 × 10⁻⁶. -5 MPa;

[0121] From the above two equations, we can obtain:

[0122]

[0123] Total sound pressure level L pt With the sound pressure level L of each sound source pi The relationship between them can be represented as:

[0124]

[0125] Let the two sound pressure levels be L1 and L2. If the total sound pressure level L and one of the sound pressure levels L1 are known, then the other sound pressure level L2 can be calculated:

[0126]

[0127] By performing one-third harmonic processing on the total noise and mechanical noise of the internal combustion engine, the noise value corresponding to each center frequency can be obtained. The sound level is decomposed by using the total noise value and mechanical noise value at that center frequency to obtain the combustion noise spectrum of the internal combustion engine.

[0128] S3: Combustion noise signal decomposition based on EFD;

[0129] The dramatic change in gas pressure inside the cylinder is the root cause of combustion noise in internal combustion engines, according to EFD-based combustion noise signal decomposition. There are three reasons for this from a mechanistic perspective.

[0130] The sudden pressure change inside the cylinder causes the cylinder and all its connected components to be subjected to strong impact dynamic loads, forming instantaneous excitation, which is transmitted to the entire internal combustion engine through the cylinder head, cylinder liner, piston connecting rod mechanism, crankshaft and its support, resulting in structural vibration of each component.

[0131] During the ignition delay period, the fuel injected into the combustion chamber ignites at several points in the combustion chamber when the combustion conditions are ripe. As a result, the local pressure at the initial ignition point rises sharply, and the flame spreads to the adjacent area. At the same time as the flame spreads, pressure waves with impact characteristics also spread. When these pressure waves encounter the combustion chamber wall, they are reflected. The repeated reflections of the pressure waves in the cylinder generate high-frequency vibrations in the gas. These high-frequency vibrations are maintained for a considerable period of time during the expansion stroke, thereby radiating high-frequency noise.

[0132] The gas pressure inside the cylinder changes periodically under stable operating conditions. By performing a Fourier transform on the cylinder pressure curve, a frequency domain pressure amplitude spectrum with the number of internal combustion engine working cycles per second as the fundamental frequency can be obtained. This constitutes the gas pressure vibration excitation, which will excite the forced vibration of the components connected inside the internal combustion engine cylinder.

[0133] The Fourier spectrum of the signal f(t) to be decomposed is obtained by Fourier transform.

[0134]

[0135] Using improved Fourier spectrum segmentation techniques to The sequence is divided and sorted, and the frequencies corresponding to the first N maximum values ​​in the sorted sequence are [Ω1, Ω2, ... Ω]. N The boundaries of the segments are defined by ω. n Sure:

[0136]

[0137] In the formula, Represents Ω n To Ω n+1 The Fourier spectrum amplitudes between;

[0138] Based on the obtained ω n Establish a zero-phase filter bank

[0139]

[0140] use Obtain the filtered signal

[0141]

[0142] Through calculation in step four The inverse Fourier transform yields different intrinsic mode functions (IMFs) and f. n (t):

[0143]

[0144] S4: Based on SST, IMF time-frequency analysis is used to obtain the intrinsic mode functions (IMFs) of the decomposed noise components;

[0145] The specific details of the SST-based IMF time-frequency analysis are as follows:

[0146] The time-frequency characteristics of the IMF are obtained by using the synchronous wavelet compression transform (SST). The synchronous wavelet compression transform is based on the CWT and compresses and rearranges the complex coefficient spectrum of the CWT along the scale direction, which improves the resolution of the CWT while maintaining its complete mathematical invertibility.

[0147] Based on the time-frequency analysis results, the main excitation of internal combustion engine combustion noise can be preliminarily determined;

[0148] Further analysis of the separated combustion noise is based on prior knowledge of internal combustion engines, the coherence of cylinder pressure and noise, and the results of the reverse drag test.

[0149] The main excitations for combustion noise in internal combustion engines include drastic changes in cylinder pressure, high-frequency vibrations of gas, and gas pressure vibration excitations.

[0150] The above are merely preferred embodiments 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. A method for identifying combustion noise in internal combustion engines based on EFD-SST, which utilizes EFD technology to adaptively decompose the combustion noise signal of an internal combustion engine and combines it with SST to identify each component, characterized in that... Includes the following steps: S1: Test the noise signals of the internal combustion engine at 5 points under standard conditions and reverse towing conditions respectively. The standard condition is 2000 r / min and the reverse towing condition is 2000 r / min. S2: The total noise and mechanical noise of the internal combustion engine are separated by using the reverse drag method. One-third harmonic processing is performed to obtain the noise value corresponding to each center frequency. The combustion noise spectrum of the internal combustion engine is obtained by using the total noise value and mechanical noise value at the center frequency. S3: Based on EFD technology, the combustion noise signal of the internal combustion engine is decomposed; S4: Based on SST, IMF time-frequency analysis technology is used to obtain the intrinsic mode functions (IMFs) of the decomposed components of noise; The separation of mechanical noise and combustion noise specifically includes: In the method of noise separation using the reverse dragging method, the overall noise includes combustion noise and mechanical noise. The experimentally measured reverse dragging noise is regarded as mechanical noise. The experimentally measured overall noise and reverse dragging noise are used to calculate the internal combustion engine combustion noise using the sound level superposition method. For n independent sound sources existing simultaneously, based on the superposition of energy and the relationship between sound power W and sound pressure P: The total sound pressure p is obtained. t The calculation formula is: In the formula, The sound pressure level of each sound source; According to the definition of sound pressure level, the expression for total sound pressure level is: In the formula, p0 is the reference sound pressure level, which is usually adopted. ; From the above two equations, we can obtain: Total sound pressure level L pt With the sound pressure level L of each sound source pi The relationship between them is represented as follows: Let the two sound pressure levels be L1 and L2. If the total sound pressure level L and one of the sound pressure levels L1 are known, then the other sound pressure level L2 can be calculated: By performing one-third harmonic processing on the total noise and mechanical noise of the internal combustion engine, the noise value corresponding to each center frequency can be obtained. The sound level is then decomposed using the total noise value and mechanical noise value at that center frequency to obtain the combustion noise spectrum of the internal combustion engine.

2. The method for identifying internal combustion engine combustion noise based on EFD-SST according to claim 1, characterized in that, The sampling frequency of the noise measurement point is 20480Hz. The noise test is conducted in a semi-anechoic chamber with dimensions of 11.2m×9.5m×6.0m, a cutoff frequency of 100Hz, and a background noise of 25dB.

3. The method for identifying internal combustion engine combustion noise based on EFD-SST according to claim 2, characterized in that, The noise measurement equipment includes an AVL internal combustion engine test system, an LMS Test.lab 32-bit acoustic vibration test system, a BSWA acoustic microphone, and an accelerometer for the PCB.

4. The internal combustion engine combustion noise identification method based on EFD-SST according to claim 1, wherein the adaptive decomposition of the combustion noise signal based on EFD technology includes: The signal to be decomposed is obtained by Fourier transform. Fourier spectrum : Using improved Fourier spectrum segmentation techniques to The sequence is divided and sorted, and the frequencies corresponding to the top N maximum values ​​in the sorted sequence are... The boundaries of the segments are determined by Sure: In the formula, express arrive The Fourier spectrum amplitudes between; According to the obtained Establish a zero-phase filter bank : use Obtain the filtered signal : Through calculation in step four The inverse Fourier transform yields different intrinsic mode functions (IMFs) and : 。 5. The internal combustion engine combustion noise identification method based on EFD-SST according to claim 1, wherein the IMF time-frequency analysis based on SST specifically comprises: The time-frequency characteristics of the IMF are obtained by using synchronous wavelet compression transform (SST). The synchronous wavelet compression transform is based on CWT, and the complex coefficient spectrum of CWT is compressed and rearranged along the scale direction. Based on the time-frequency analysis results, the main excitation of internal combustion engine combustion noise can be preliminarily determined; Based on prior knowledge of internal combustion engines, the coherence of cylinder pressure and noise, and the results of the reverse drag test, the separation of combustion noise is further determined.

6. The internal combustion engine combustion noise identification method based on EFD-SST according to claim 1 can effectively identify the main excitations of internal combustion engine combustion noise, including sudden changes in cylinder pressure, high-frequency vibration of gas, and gas pressure vibration excitation.

Images