Method and system for monitoring running state of diesel engine in real time

Abstract

The invention relates to the technical field of electronic data processing, and discloses a diesel engine operation state real-time monitoring method and system, and the method comprises the steps: obtaining a vibration signal of a measurement point on the surface of an engine body as an original signal; modal decomposition is carried out to obtain a final intrinsic mode function, and the decomposition process comprises the steps that multiple times of lumped operation are executed, in each time of lumped operation, an auxiliary noise signal is added into a to-be-decomposed original signal, then the screening iteration process is executed repeatedly, and a set of temporary intrinsic mode functions are obtained through separation in sequence; performing weighted average on the temporary intrinsic mode functions of the same order obtained in all the lumped operations to obtain a final intrinsic mode function of the order; hilbert transform is carried out on the intrinsic mode function, joint distribution of instantaneous energy and instantaneous frequency of the intrinsic mode function is constructed, and statistical and geometric morphological features are extracted to form feature vectors; according to the method, modal aliasing can be inhibited, noise residues are eliminated, and the accuracy of a running state monitoring result is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic data processing, and in particular to a method and system for real-time monitoring of the operating status of a diesel engine. Background Art

[0002] Diesel engines, as key power equipment, are used in transportation, construction machinery, ship propulsion, and power generation. Real-time and accurate monitoring of their operating status is crucial for ensuring safe and reliable equipment operation, preventing sudden failures, and reducing maintenance costs. Vibration signal analysis, a non-invasive, information-rich condition monitoring technology, has become a mainstream research direction in this field. Diesel engine vibration signals are typically nonlinear and non-stationary, containing a wealth of operating status information. Traditional signal processing methods, such as the Fourier transform, are only suitable for analyzing stationary signals and have difficulty revealing the time-varying characteristics of signals. While the wavelet transform has time-frequency analysis capabilities, its performance is limited by the preselected wavelet basis functions and lacks adaptability, which can lead to energy leakage or feature misjudgment. To overcome these limitations, empirical mode decomposition (EMD) has been proposed as an adaptive signal processing method. It decomposes complex signals into a series of intrinsic mode functions (IMFs) based on the signal's inherent time scale characteristics, demonstrating unique advantages in processing non-stationary signals.

[0003] However, traditional EMD methods suffer from the "modal aliasing" problem, whereby a single IMF contains components of different time scales, or components of the same scale are dispersed across different IMFs. This severely impacts the accuracy of subsequent feature extraction and the interpretability of physical meaning. To suppress modal aliasing, ensemble empirical mode decomposition (EEMD) and its improved algorithms have been proposed. EEMD adds Gaussian white noise to the original signal multiple times, leveraging the uniform distribution of the white noise spectrum to aid signal decomposition. The decompositions are then averaged to offset the added noise. Although EEMD has alleviated modal aliasing to a certain extent, it still has several shortcomings: First, when the number of lumped averages is limited, the auxiliary white noise cannot be completely neutralized, resulting in residual noise in the decomposition result, affecting the purity of the IMF component; second, its core screening iterative process still uses the EMD framework, and has defects in the selection of extreme points, envelope construction and the setting of iteration termination conditions. For example, it is sensitive to noise spikes, interpolation is prone to overshoot and undershoot, and the iteration stopping criterion is single. These problems limit the accuracy and efficiency of the decomposition; third, the huge amount of computation limits its use in online monitoring applications that require high real-time performance. Summary of the Invention

[0004] The present invention provides a method and system for real-time monitoring of the operating status of a diesel engine to solve the problem in the prior art that modal aliasing and residual noise lead to inaccurate operating status monitoring results.

[0005] In a first aspect, the method for real-time monitoring of the operating status of a diesel engine of the present invention comprises the following steps: The vibration signal of the measuring point on the surface of the machine body is obtained as the original signal; the original signal is subjected to ensemble empirical mode decomposition to obtain the final intrinsic mode function. The ensemble empirical mode decomposition process includes: performing multiple lumping operations, adding an auxiliary noise signal matched with statistical characteristics to the original signal to be decomposed in each lumping operation, repeating the screening iterative process on the signal after adding the auxiliary noise signal, and successively separating a group of temporary intrinsic mode functions; the screening iterative process includes: a. adaptively screening local extreme points according to the local and global statistical characteristics of the signal to be processed; b. using the local adaptive interpolation method to construct the upper and lower envelopes based on the retained extreme points; c. calculating the mean of the upper and lower envelopes and subtracting them from the signal to be processed. The mean is repeated from a to c until the preset termination condition is met. The termination condition comprehensively considers the energy stability of the current component and its orthogonality with the separated temporary intrinsic mode function; the temporary intrinsic mode functions of the same order obtained in all lumped operations are weighted averaged to obtain the final intrinsic mode function of the order, and the final intrinsic mode functions of all orders constitute the final intrinsic mode function set, where the weights are determined according to the signal-to-noise ratio or stability index of each temporary intrinsic mode function; one or more selected intrinsic mode functions in the final intrinsic mode function set are Hilbert transformed to construct the joint distribution of their instantaneous energy and instantaneous frequency, and the statistical and geometric features of the joint distribution are extracted to form a feature vector.

[0006] Preferably, the adding of the auxiliary noise signal matched with the statistical characteristics to the original signal to be decomposed includes: generating a set of Gaussian white noise signals of the same length as the original signal to be decomposed, and adjusting the amplitude of the Gaussian white noise signal so that its standard deviation is a preset ratio of the global standard deviation of the original signal to be decomposed, and the preset ratio is 0.1 to 0.3; and then adding the adjusted Gaussian white noise signal to the original signal to be decomposed point by point.

[0007] Preferably, the method of adaptively screening local extreme points based on the local and global statistical characteristics of the signal to be processed includes: calculating the global standard deviation σ of the signal to be processed g And set the global screening threshold T g , T g =α σ g ; For each extreme point, calculate the local standard deviation σ in its neighborhood window l And set the local screening threshold T l=β σ l ; Only when the absolute amplitude of the extreme point is greater than T g and T l , the extreme point is retained; where α is a coefficient between 0.3 and 0.7, and β is a coefficient between 1.0 and 1.5.

[0008] Preferably, the termination condition includes: setting the energy convergence threshold ε and the orthogonality index threshold ξ; in the screening iteration process, if the intermediate component obtained in the current iteration is h k (t), the intermediate component obtained in the previous iteration is h k-1 (t), then calculate the standard deviation criterion SD=Σ[(h k (t)-h k-1 (t)) 2 ] / Σ[h k-1 (t) 2 ]; When SD is less than ε, and h k When the mutual correlation coefficient of (t) with each separated temporary intrinsic mode function in the current lumped operation is less than ξ, the iteration is terminated.

[0009] Preferably, the weighted averaging of the temporary intrinsic modal functions of the same order obtained in all lumped operations includes: calculating the instantaneous frequency sequence of each temporary intrinsic modal function of the same order through Hilbert transform; calculating the standard deviation of the instantaneous frequency sequence, and using the inverse of the standard deviation as a weight to characterize its stability; normalizing the weights of all temporary intrinsic modal functions of the same order, and performing weighted summation on the corresponding temporary intrinsic modal functions accordingly to obtain a final intrinsic modal function of the order.

[0010] Preferably, the statistical and geometric features include: statistical moments of the one-dimensional distribution obtained by projecting the joint distribution onto the instantaneous energy axis and the instantaneous frequency axis, wherein the statistical moments are selected from at least two of the mean, standard deviation, skewness, and kurtosis.

[0011] Preferably, the statistical and geometric features further include: the area and / or perimeter of a two-dimensional convex hull formed by the set of points in the joint distribution on the energy-frequency plane, and the centroid coordinates of the joint distribution on the energy-frequency plane.

[0012] Preferably, the local adaptive interpolation method is a piecewise cubic Hermite interpolation polynomial method.

[0013] Preferably, the obtaining of vibration signals from measuring points on the engine body surface as original signals includes: installing a piezoelectric acceleration sensor on the engine body near the cylinder head, collecting vibration acceleration signals through a data acquisition card, and using the collected time series signals as original signals.

[0014] In a second aspect, the diesel engine operating status real-time monitoring system of the present invention includes a memory and a processor. The memory stores computer instructions. When the processor executes the computer instructions, the above-mentioned diesel engine operating status real-time monitoring method is implemented.

[0015] The beneficial effects of the present invention are as follows: in the screening and iterative process of decomposition, the present invention introduces an extreme point discrimination mechanism based on the overall and local characteristics of the signal, which effectively avoids the interference of noise spikes; and adopts an optimized interpolation algorithm to construct the envelope, reducing waveform distortion. At the same time, the iterative termination condition combines the dual indicators of energy stability and modal orthogonality, ensuring that the physical meaning of each decomposed inherent modal function is clear and independent of each other, and suppressing modal aliasing. By weighted averaging the temporary modes obtained by multiple decompositions, the auxiliary noise residue is further eliminated, and a final modal component with higher purity is obtained. On this basis, the joint distribution of instantaneous energy and frequency is constructed and its morphological characteristics are extracted to express the subtle changes in the operating status of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a method for real-time monitoring of the operating status of a diesel engine provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0018] like Figure 1 As shown, an embodiment of the method for real-time monitoring of the operating status of a diesel engine provided by the present invention includes the following steps: S1. Obtain vibration signals of measuring points on the surface of a diesel engine body as original signals.

[0019] Specifically, a piezoelectric acceleration sensor is installed on the diesel engine body near the cylinder head, and a data acquisition card such as the NI 9234 model board is used to collect vibration acceleration signals at a sampling frequency of 25.6 kHz, and the collected time series signals are used as original signals.

[0020] S2. Perform an improved collective empirical mode decomposition on the original signal to obtain a set of final intrinsic mode functions. The decomposition process includes: performing multiple lumping operations, adding an auxiliary noise signal matched with statistical characteristics to the original signal to be decomposed in each lumping operation, and repeatedly performing a screening iterative process on the signal after adding noise to obtain a set of temporary intrinsic mode functions.

[0021] Specifically, the total number of lumped operations is set to, for example, 100 times. In the i-th operation, a Gaussian white noise sequence with the same length as the current signal to be decomposed is generated, and the amplitude of the noise sequence is adjusted so that its standard deviation is 0.2 times the standard deviation of the signal to be decomposed. Then, the noise sequence is added to the signal to be decomposed to form a noisy signal, and screening iterations are performed on this noisy signal.

[0022] The screening iterative process is used to isolate a single temporary intrinsic mode function, which includes: a. Adaptively filter local extreme points based on the local and global statistical characteristics of the signal to be processed. Specifically, find all local maxima and minima in the signal, then calculate the overall standard deviation of the signal and set a global threshold equal to 0.1 times the standard deviation. For each extreme point, calculate the local standard deviation within a small window of its neighborhood, such as 50 data points, and set a local threshold. Only when the absolute value of an extreme point is greater than both the global threshold and its corresponding local threshold is it retained as a valid extreme point.

[0023] b. Using a local adaptive interpolation method, upper and lower envelopes are constructed based on the retained extreme points. Specifically, a piecewise cubic Hermite interpolation polynomial is used to connect all retained maximum points to form the upper envelope, and then connect all retained minimum points to form the lower envelope. This method ensures the monotonicity of the interpolation curve between extreme points, avoiding the overshoot and oscillation that can occur with traditional spline interpolation.

[0024] c. Calculate the mean of the upper and lower envelopes and subtract the mean from the signal to be processed, repeat the iterations from a to c until the preset termination condition is met. The termination condition comprehensively considers the energy stability of the current component and its orthogonality with the separated temporary intrinsic mode function. Specifically, the upper and lower envelopes are added point by point and divided by two to obtain the mean envelope, which is subtracted from the signal to be processed to obtain a candidate component. Calculate the standard deviation ratio between the candidate component and the candidate component obtained in the previous iteration, and calculate the orthogonality index between the candidate component and all the separated final intrinsic mode functions. When the standard deviation ratio is less than the energy convergence threshold, such as 0.01, and the orthogonality index is less than the orthogonality threshold, such as 0.05, the iteration terminates, and the current candidate component is confirmed as a temporary intrinsic mode function.

[0025] S3. Perform weighted averaging on the temporary intrinsic modal functions of the same order obtained in all lumped operations to obtain the final intrinsic modal function of the order. The final intrinsic modal functions of all orders constitute the final intrinsic modal function set, where the weights are determined according to the signal-to-noise ratio or stability index of each temporary intrinsic modal function.

[0026] For example, after 100 lumped operations are completed, for the first order, 100 temporary first-order intrinsic mode functions are obtained; for each of these 100 temporary mode functions, its Pearson correlation coefficient with the original signal is calculated as its weight; then each temporary mode function is multiplied by its corresponding normalized weight and summed to obtain the final first-order intrinsic mode function; then this process is repeated for subsequent orders.

[0027] S4. Performing a Hilbert transform on one or more selected intrinsic mode functions in the final intrinsic mode function set to construct a joint distribution of their instantaneous energy and instantaneous frequency, and extracting statistical and geometric features of the joint distribution to form a feature vector for characterizing and monitoring the operating status of the diesel engine.

[0028] Specifically, one or more final intrinsic modal functions are selected based on the energy proportion or the correlation with the fault characteristic frequency. Each selected modal function is subjected to Hilbert transform to obtain its analytical signal, thereby calculating its instantaneous amplitude and instantaneous frequency that change with time. The instantaneous energy is the square of the instantaneous amplitude. This forms a three-dimensional distribution diagram of energy-frequency-time. Features are extracted from the distribution, such as the mean and variance of the instantaneous energy, the mean and variance of the instantaneous frequency, the kurtosis and skewness of the energy distribution, and the area and centroid position of the main energy accumulation area. These numerical features are combined into a feature vector and input into a pre-trained support vector machine classifier to output the specific operating status of the diesel engine, such as normal, misfire or bearing wear.

[0029] In an optional embodiment, the adding of an auxiliary noise signal with matched statistical characteristics to the original signal to be decomposed includes: generating a set of Gaussian white noise signals of the same length as the original signal to be decomposed, and adjusting the amplitude of the Gaussian white noise signal so that its standard deviation is a preset proportion of the global standard deviation of the original signal to be decomposed, and the preset proportion is 0.1 to 0.3; and then adding the adjusted Gaussian white noise signal to the original signal to be decomposed point by point.

[0030] The addition of Gaussian white noise can help distinguish signal components with similar amplitudes but large frequency differences, or intermittent signals with similar frequencies, to ensure the accuracy of subsequent decomposition. The statistical characteristic matching in this process is the key, which ensures that the added noise can not only play a role in auxiliary decomposition, but also does not excessively submerge the useful information of the original signal. For example, assuming that the length of the vibration signal to be decomposed is 2048 data points, its global standard deviation is calculated to be 2.0 millivolts. The preset ratio is selected as 0.2, which is within the preferred range of 0.1 to 0.3. Therefore, the target standard deviation of the auxiliary noise signal to be generated is 0.4 millivolts. Generate a Gaussian white noise sequence with the same length of 2048 points, and adjust its standard deviation to 0.4 millivolts precisely through linear scaling. Add the adjusted noise sequence to each data point of the original vibration signal one by one to form a new signal to be processed.

[0031] In an optional embodiment, the method of adaptively screening local extreme points based on the local and global statistical characteristics of the signal to be processed includes: calculating the global standard deviation σ of the signal to be processed g And set the global screening threshold T g , T g =α σ g ; For each extreme point, calculate the local standard deviation σ in its neighborhood window l And set the local screening threshold T l =β σ l ; Only when the absolute amplitude of the extreme point is greater than T g and T l , the extreme point is retained; where α is a coefficient between 0.3 and 0.7, and β is a coefficient between 1.0 and 1.5.

[0032] The dual threshold screening mechanism can intelligently identify and retain real signal fluctuations while effectively suppressing noise interference. Global threshold T g It is mainly used to filter out the low-amplitude background noise that is prevalent in the entire signal, ensuring that only fluctuations with a certain energy are considered. l It is dynamically adjusted according to the local complexity of the signal. In the signal stationary area, the local standard deviation σ l Very small, T l is also small accordingly, which can preserve weak but real extreme points; in areas where the signal fluctuates violently, σ l Very big, T l It also increases, which can effectively eliminate the pseudo extreme points caused by noise and avoid over-decomposition. Assume that the global standard deviation σ of a signal g is 5 units, and α is set to 0.6, then the global threshold T gFor a maximum point with a value of 4.5 in the signal, its absolute amplitude is greater than T g , satisfying the first condition. Examine a neighborhood window around the point, for example, 10 data points before and after, and calculate the local standard deviation σ within this window l is 2 units. Set β to 1.2, then the local threshold T l is 2.4 units. Since the amplitude of the maximum point is 4.5, which is greater than both the global threshold of 3 and the local threshold of 2.4, the extreme point is retained as a valid point for constructing the envelope. On the contrary, if there is another extreme point with an amplitude of 3.5, although it is greater than the global threshold, the neighborhood where it is located fluctuates violently, and the local standard deviation σ l up to 4, resulting in a local threshold T l The extreme point is 4.8, and is eliminated because its amplitude is less than the local threshold.

[0033] In an optional embodiment, the termination condition includes: setting the energy convergence threshold ε and the orthogonality index threshold ξ; in the screening iteration process, if the intermediate component obtained in the current iteration is h k (t), the intermediate component obtained in the previous iteration is h k-1 (t), then calculate the standard deviation criterion SD=Σ[(h k (t)-h k-1 (t)) 2 ] / Σ[h k-1 (t) 2 ]; When SD is less than ε, and h k When the mutual correlation coefficient of (t) with each separated temporary intrinsic mode function in the current lumped operation is less than ξ, the iteration is terminated.

[0034] The termination condition ensures the convergence of the screening iterative process and the validity of the decomposition results. The standard deviation criterion SD is essentially the relative change in signal energy between two iterations. When this value is less than the preset energy convergence threshold ε, it indicates that the screening process has stabilized and continued iteration will not significantly change the component, so it can be stopped. This ensures the stability of the intrinsic mode function. At the same time, the second condition, the orthogonality check, avoids serious mode aliasing between the newly extracted components and the extracted components. For example, the energy convergence threshold ε is set to 0.001 and the orthogonality index threshold ξ is set to 0.05. After the kth iteration, the calculated SD value is 0.0008, which is less than ε and meets the convergence condition. At this time, it is assumed that two temporary intrinsic mode functions timf1 and timf2 have been separated in this lumped operation. It is necessary to calculate the current component h k The cross-correlation coefficient between (t) and timf1 is 0.03, and the cross-correlation coefficient between (t) and timf2 is 0.04. Since both 0.03 and 0.04 are less than the orthogonality threshold ξ, it indicates that hk (t) is sufficiently independent from the previously separated components. Only when these two conditions are met at the same time, the screening iteration is officially terminated and h k (t) is output as a temporary intrinsic mode function.

[0035] In an optional embodiment, a weighted average is performed on the temporary intrinsic modal functions of the same order obtained in all lumped operations, including: calculating the instantaneous frequency sequence of each temporary intrinsic modal function of the same order through Hilbert transform; calculating the standard deviation of the instantaneous frequency sequence, and using the inverse of the standard deviation as a weight to characterize its stability; normalizing the weights of all temporary intrinsic modal functions of the same order, and performing weighted summation on the corresponding temporary intrinsic modal functions accordingly to obtain a final intrinsic modal function of the order.

[0036] A weighting strategy based on frequency stability is used to optimize and integrate the results of multiple noise-assisted decompositions. Because the white noise added each time is random, the resulting temporary intrinsic mode functions of the same order will vary slightly. An ideal intrinsic mode function should have a relatively stable instantaneous frequency. Therefore, the smaller the standard deviation of the instantaneous frequency series, the smaller the frequency fluctuation of that component, the clearer the physical meaning, and the higher the quality. Using the inverse of this standard deviation as the weight achieves the goal of giving greater voice to high-quality components.

[0037] For example, after completing 10 lumped operations, 10 first-order temporary intrinsic mode functions are obtained; for the i-th function, the standard deviation of its instantaneous frequency is calculated as σ i Assume that the frequency standard deviations of the first three functions are σ1=2 Hz, σ2=5 Hz, and σ3=1.5 Hz. Then their corresponding original weights are w1=1 / 2=0.5, w2=1 / 5=0.2, and w3=1 / 1.5≈0.67. Add up all 10 original weights to get the total weight W total , normalize each original weight by dividing it by the total weight to obtain the final weighting coefficient. The third function with the most stable frequency accounts for the largest proportion in the final first-order natural mode function synthesis, while the second function with larger frequency fluctuations accounts for a smaller proportion.

[0038] In an optional embodiment, the statistical and geometric features include: the statistical moments of the one-dimensional distribution obtained by projecting the joint distribution onto the instantaneous energy axis and the instantaneous frequency axis, and the statistical moments are selected from at least two of the mean, standard deviation, skewness, and kurtosis.

[0039] The decomposed intrinsic mode functions are converted into numerical features that can be used for quantitative analysis and machine learning models. Through the Hilbert transform, each intrinsic mode function can be mapped into a three-dimensional space of energy, frequency, and time. The joint distribution of these functions on the energy-frequency plane reveals the intrinsic state of the signal. Projecting this two-dimensional joint distribution onto the instantaneous frequency axis yields a one-dimensional probability distribution with respect to frequency. Calculating the mean of this distribution reveals the center frequency of the modal component, the standard deviation indicates its frequency bandwidth, the skewness describes the symmetry of the frequency distribution, and the kurtosis reflects the degree of centralization or dispersion of the frequency distribution.

[0040] Similarly, projecting the joint distribution onto the instantaneous energy axis yields a one-dimensional probability distribution of energy. Calculating its mean reveals the average energy intensity of this mode, while the standard deviation reflects the magnitude of energy fluctuations. For example, when a gearbox is operating normally, the frequency distribution mean of a key intrinsic mode function may stabilize at 100 Hz, with a standard deviation of 5 Hz and a skewness close to 0. When an incipient crack fault occurs, the frequency mean may remain unchanged, but the standard deviation may increase to 15 Hz, and the skewness becomes positive, reflecting the emergence of nonlinear modulation. These statistical moments together constitute a multidimensional feature vector that can sensitively capture changes in equipment status.

[0041] In an optional embodiment, the statistical and geometric features also include: the area and / or perimeter of the two-dimensional convex hull formed by the set of points in the joint distribution on the energy-frequency plane, and the centroid coordinates of the joint distribution on the energy-frequency plane.

[0042] On the energy-frequency plane, all data points form a point cloud. The two-dimensional convex hull of this point cloud is the smallest convex polygon that encloses all the data points. The area of ​​this convex hull directly reflects the combined magnitude of the signal's energy and frequency variation. For a healthy and stable system, the energy-frequency point cloud corresponding to the modal components is typically clustered, with a small convex hull area. When a system fails or experiences an abnormality, new frequency components or sudden energy changes may appear in the signal, causing the point cloud to become dispersed and the convex hull area to increase significantly. Furthermore, the centroid coordinates of the point cloud indicate the center of gravity of the entire distribution on the energy-frequency plane. For example, when a bearing is operating normally, the center of mass of its first mode's energy-frequency distribution may be located at coordinates , with an energy of 0.1 joule and a frequency of 30 Hz. When the bearing wears, friction increases, and the speed fluctuates slightly, the center of mass may drift to a position with an energy of 0.3 joule and a frequency of 35 Hz. By monitoring changes in the convex hull area and the center of mass coordinates, changes in the system's dynamic characteristics can be very intuitively detected.

[0043] The implementation principle of the real-time monitoring method for the operating status of a diesel engine according to an embodiment of the present invention is as follows: in the screening and iterative process of decomposition, the present invention effectively avoids the interference of noise spikes by introducing an extreme point discrimination mechanism based on the overall and local characteristics of the signal; at the same time, an optimized interpolation algorithm is used to construct the envelope to reduce waveform distortion. Moreover, the iterative termination condition combines the dual indicators of energy stability and modal orthogonality, ensuring that the physical meaning of each decomposed inherent modal function is clear and independent of each other, thereby suppressing modal aliasing. In addition, by weighted averaging the temporary modes obtained through multiple decompositions, the auxiliary noise residue is further eliminated, and a final modal component with higher purity is obtained. On this basis, the joint distribution of instantaneous energy and frequency is constructed, and its morphological characteristics are extracted, which can better express the subtle changes in the operating status of the equipment.

[0044] An embodiment of the diesel engine operating status real-time monitoring system provided by the present invention includes a memory and a processor. The memory stores computer instructions. When the processor executes the computer instructions, the diesel engine operating status real-time monitoring method in the above embodiment is implemented.

[0045] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for real-time monitoring of the operating status of a diesel engine, characterized in that: include: Obtain vibration signals of measuring points on the body surface as original signals; The original signal is subjected to ensemble empirical mode decomposition to obtain the final intrinsic mode function. The ensemble empirical mode decomposition process includes: performing multiple lumping operations, adding an auxiliary noise signal matched with statistical characteristics to the original signal to be decomposed in each lumping operation, repeating the screening iterative process on the signal after adding the auxiliary noise signal, and successively separating a set of temporary intrinsic mode functions; the screening iterative process includes: a. adaptively screening local extreme points according to the local and global statistical characteristics of the signal to be processed; b. using a local adaptive interpolation method to construct upper and lower envelopes based on the retained extreme points; c. calculating the mean of the upper and lower envelopes and subtracting the mean from the signal to be processed, and repeating a to c. Iterate until the preset termination condition is met. The termination condition comprehensively considers the energy stability of the current component and its orthogonality with the separated temporary intrinsic mode function; the temporary intrinsic mode functions of the same order obtained in all lumped operations are weighted averaged to obtain the final intrinsic mode function of the order, and the final intrinsic mode functions of all orders constitute the final intrinsic mode function set, where the weights are determined according to the signal-to-noise ratio or stability index of each temporary intrinsic mode function; perform Hilbert transform on one or more selected intrinsic mode functions in the final intrinsic mode function set, construct the joint distribution of their instantaneous energy and instantaneous frequency, and extract the statistical and geometric features of the joint distribution to form a feature vector.

2. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The method of adding an auxiliary noise signal having statistical characteristics matched to the original signal to be decomposed includes: generating a set of Gaussian white noise signals of equal length to the original signal to be decomposed, and adjusting the amplitude of the Gaussian white noise signal so that its standard deviation is a preset ratio of the global standard deviation of the original signal to be decomposed, wherein the preset ratio is 0.1 to 0.3; and then adding the adjusted Gaussian white noise signal to the original signal to be decomposed point by point.

3. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The method of adaptively screening local extreme points based on the local and global statistical characteristics of the signal to be processed includes: calculating the global standard deviation σ of the signal to be processed g And set the global screening threshold T g , T g =α σ g ; For each extreme point, calculate the local standard deviation σ in its neighborhood window l And set the local screening threshold T l =β σ l ; Only when the absolute amplitude of the extreme point is greater than T g and T l , the extreme point is retained; where α is a coefficient between 0.3 and 0.7, and β is a coefficient between 1.0 and 1.

5.

4. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The termination conditions include: setting the energy convergence threshold ε and the orthogonality index threshold ξ; in the screening iteration process, if the intermediate component obtained in the current iteration is h k (t), the intermediate component obtained in the previous iteration is h k-1 (t), then calculate the standard deviation criterion SD=Σ[(h k (t)-h k-1 (t)) 2 ] / Σ[h k-1 (t) 2 ]; When SD is less than ε, and h k When the mutual correlation coefficient of (t) with each separated temporary intrinsic mode function in the current lumped operation is less than ξ, the iteration is terminated.

5. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The weighted averaging of temporary intrinsic modal functions of the same order obtained from all lumped operations includes: calculating the instantaneous frequency sequence of each temporary intrinsic modal function of the same order through Hilbert transform; calculating the standard deviation of the instantaneous frequency sequence, and using the inverse of the standard deviation as a weight to characterize its stability; normalizing the weights of all temporary intrinsic modal functions of the same order, and performing weighted summation on the corresponding temporary intrinsic modal functions based on the normalized weights to obtain a final intrinsic modal function of the order.

6. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The statistical and geometric features include: statistical moments of a one-dimensional distribution obtained by projecting the joint distribution onto the instantaneous energy axis and the instantaneous frequency axis, wherein the statistical moments are selected from at least two of the mean, standard deviation, skewness, and kurtosis.

7. The method for real-time monitoring of the operating status of a diesel engine according to claim 6, characterized in that: The statistical and geometric features also include: the area and / or perimeter of a two-dimensional convex hull formed by the point set in the joint distribution on the energy-frequency plane, and the centroid coordinates of the joint distribution on the energy-frequency plane.

8. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The local adaptive interpolation method is a piecewise cubic Hermite interpolation polynomial method.

9. The method for real-time monitoring of the operating status of a diesel engine according to claim 1, characterized in that: The method of obtaining the vibration signal of the measuring point on the engine body surface as the original signal includes: installing a piezoelectric acceleration sensor on the engine body near the cylinder head, collecting the vibration acceleration signal through a data acquisition card, and using the collected time series signal as the original signal.

10. A diesel engine operating status real-time monitoring system, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer instructions, and when the processor executes the computer instructions, the method for real-time monitoring of the operating status of a diesel engine as claimed in any one of claims 1 to 9 is implemented.

Images