Vibration signal self-adaptive impact interception method
By calculating the noise effective value and zeroing processing, the impact signal in the vibration signal is adaptively intercepted, and the problem of intercepting incomplete and noisy signals in the prior art is solved, and complete signal interception and fault diagnosis are achieved.
Patent Information
- Application Number
- CN202410979384.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
AI Technical Summary
In the existing vibration signal impact interception technology, the threshold method causes the intercepted impact signal to be incomplete, and the window law may contain noisy signals, lacking the single interception and retention function of impact.
By calculating the mean and variance of the noise effective value, obtain the absolute peak value of the vibration signal and its position, and find a data point greater than the mean of the noise effective value plus 3 times the standard deviation before and after the peak position, set the zero processing to intercept the impact signal, repeat Up to the peak value less than the noise effective value plus 3 times the standard deviation.
Adaptive interception of complete vibration shock signals is achieved, ensuring the integrity of the intercepted signal and effectively diagnosing internal combustion engine misfire faults.
Smart Images

Figure CN118940107A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vibration signal impact interception, and in particular to a vibration signal adaptive impact interception method. Background Art
[0002] The impact is often caused by a short-term vibration of parts due to external stimulation. In the equipment status monitoring, it is manifested as the impact of the vibration signal. The vibration signal with an amplitude significantly higher than the basic vibration part and a shorter duration is considered to be an impact signal. The impact often contains information about the working status of the moving parts and when a fault occurs, so it is particularly important to intercept and analyze the impact of the vibration signal. Especially in the operation of reciprocating machinery, the impact of the vibration signal is periodic. The amplitude, effective value and other parameters of the impact in the vibration signal are analyzed to perform fault diagnosis. For example, in the cylinder head vibration signal of an internal combustion engine, the ignition impact characteristic parameters are analyzed to perform misfire fault diagnosis.
[0003] At present, the threshold method and the window method are commonly used technical means in the interception of vibration signal impact. The threshold method is to set a threshold line in the global vibration signal, and the signal with a vibration amplitude exceeding the threshold is considered to be an impact; the window method is to intercept a certain length of signal at the position where the vibration signal impact may occur. However, when the threshold method is used to intercept the impact, due to the problem of setting the threshold line, only the part of the intercepted impact signal with a vibration amplitude exceeding the threshold line is retained, and the part below the threshold line is discarded, resulting in incomplete intercepted impact signals, and the intercepted impact signal is intercepted together with the global impact signal, and does not have the function of intercepting and retaining a single impact; and the impact signal intercepted by the window method often contains noise signals other than vibration impact or the intercepted impact signal is incomplete due to the problem of setting the window interception length. Summary of the invention
[0004] The main purpose of the present invention is to provide a vibration signal adaptive impact interception method to solve the problems of difficulty in setting the threshold value of the existing threshold method, incomplete intercepted vibration impact signal, and lack of impact single interception and retention function.
[0005] Another object of the present invention is to provide a vibration signal adaptive impact interception method to solve the problem that the vibration impact signal intercepted by the window method is incomplete and contains noise signals other than the impact.
[0006] In order to achieve the above object, the present invention provides the following technical solutions:
[0007] A vibration signal adaptive impact interception method is provided, which calculates the mean and variance of the effective value of noise to obtain the absolute value peak of the vibration signal and its position; obtains the two points closest to the peak position whose effective values before and after the peak position are greater than the mean of the effective value of noise plus 3 times the standard deviation, and sets the impact between the two points to zero in the original waveform; repeats the above interception method until the peak value is less than the mean of the effective value of noise plus 3 times the standard deviation.
[0008] Furthermore, the method specifically comprises:
[0009] Step 1: Calculate the sliding window vibration effective value of the noise component of the vibration signal, and calculate the mean μ and standard deviation σ of the sliding window effective value of the noise component;
[0010] Step 2: Use the sliding window method to calculate the sliding window effective value of the original vibration signal;
[0011] Step 3: Take the absolute value of the vibration signal, find the peak value and its position after taking the absolute value of the vibration signal, and compare it with the vibration signal after the sliding window effective value calculation is completed. Find the data point position closest to the peak position and greater than the noise effective value mean plus three times the standard deviation before and after the peak position. and Obtain the intercepted impact signal C1 and The signal of the section is reset to zero;
[0012] Step 4: Repeat the above method for the zeroed vibration signal to find the peak value and the data point positions before and after the peak value. and Get the intercepted impact signal C2, and then The signal of the segment is reset to zero; repeat the above steps to intercept The impulse signal Cn of the interval segment is adjusted until the peak value is less than or equal to the mean of the effective value of the noise plus three times the standard deviation.
[0013] Furthermore, in step 1 and step 2, the sliding window effective value calculation formula is as follows:
[0014]
[0015] Where s(k) is the data point at position k in the vibration signal sequence, N wl (N wl =2N w +1) is the window width, N w According to empirical settings, Y(i) is the effective value of the window at position i in the vibration signal sequence, and n is the length of the vibration signal sequence.
[0016] Furthermore, in step 1, the calculation formula of the mean μ is as follows:
[0017]
[0018] Where Yns(i) is the effective value sequence of the vibration noise sliding window.
[0019] Furthermore, in step 1, the standard deviation σ is calculated as follows:
[0020]
[0021] Where Yns(i) is the effective value sequence of the vibration noise sliding window.
[0022] Further, in step 3, the intercepted impulse signal C1 is the data point position The data between.
[0023] Further, in step 4, the intercepted impulse signal C2 is at the data point position The data between.
[0024] Compared with the prior art, the vibration signal adaptive impact interception method provided by the present invention has the following beneficial effects:
[0025] The present invention can adaptively intercept a complete vibration impact signal, and the intercepted vibration impact signal is complete.
[0026] Furthermore, the present invention can adaptively intercept the impact component in the vibration signal; at the same time, by analyzing the impact component intercepted by the present invention, it can be determined whether a misfire fault has occurred in the signal, so the present invention can effectively diagnose a misfire fault in an internal combustion engine. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for the technical description.
[0028] Figure 1 A schematic diagram of the diesel engine cylinder head vibration acceleration signal provided by the existing method;
[0029] Figure 2 A schematic diagram of the effective value of the sliding window of vibration acceleration of a diesel engine cylinder head provided by the present invention;
[0030] Figure 3 A schematic diagram of an impact signal captured by the present invention;
[0031] Figure 4 A schematic diagram of a normal diesel engine cylinder head vibration signal provided by an existing method;
[0032] Figure 5 A schematic diagram of an impact waveform captured by the present invention;
[0033] Figure 6Schematic diagram of diesel engine misfire fault vibration signal provided by existing methods;
[0034] Figure 7 This is a schematic diagram of another form of shock waveform captured by the present invention. DETAILED DESCRIPTION
[0035] The following is further explained in detail through specific implementation methods.
[0036] The present invention first calculates the effective value of the sliding window vibration of the noise component of the vibration signal, and calculates the mean μ and standard deviation σ of the effective value of the sliding window of the noise component; uses the existing sliding window method to calculate the effective value of the sliding window for the vibration signal; then takes the absolute value of the vibration signal for processing, finds the peak value and its position after the absolute value of the vibration signal is taken, and compares the vibration signal after the sliding window effective value calculation is completed, and finds the two data point positions closest to the peak position and greater than the mean value of the noise effective value plus three times the standard deviation (μ+3σ, 3σ criterion) before and after the peak position, and the signal between the two data point positions is the intercepted impact signal C1; then the signal of the section is zeroed; then the above method is repeated for the vibration signal after the zeroing process to find the peak value, and the two data point positions before and after the peak position are found, then the two data point positions are the intercepted impact signal C2, and then the signal of the section is zeroed; repeats the above method to intercept the impact signal Cn until the peak value is less than or equal to the mean value of the noise effective value plus three times the standard deviation.
[0037] Specifically, the present invention provides a vibration signal adaptive impact interception method, comprising the following steps:
[0038] Step 1: Calculate the sliding window vibration effective value of the noise component of the vibration signal, and calculate the mean μ and standard deviation σ of the sliding window effective value of the noise component.
[0039] Step 2: Use the sliding window method to calculate the sliding window effective value of the original vibration signal.
[0040] Step 3: Take the absolute value of the vibration signal, find the peak value and its position after taking the absolute value of the vibration signal, and compare it with the vibration signal after the sliding window effective value calculation is completed. Find the data point position closest to the peak position and greater than the noise effective value mean plus three times the standard deviation (μ+3σ, 3σ criterion) before and after the peak position. and Then at the data point location The data between is considered as the intercepted impulse signal C1, and The signal of the segment is reset to zero.
[0041] Step 4: Repeat the above method for the zeroed vibration signal to find the peak value and the data point positions before and after the peak value. and Then at the data point location The data between is considered as the intercepted impact signal C2, and then The signal of the segment is reset to zero; repeat the above method to intercept The impact Cn of the interval segment ends when the peak value is less than or equal to the mean of the effective value of the noise plus three times the standard deviation.
[0042] In step 1 and step 2, the calculation formula of the effective value of the sliding window is as follows:
[0043]
[0044] Where s(k) is the data point at position k in the vibration signal sequence, N wl (N wl =2N w +1) is the window width, N w According to empirical settings, Y(i) is the effective value of the window at position i in the vibration signal sequence, and n is the length of the vibration signal sequence.
[0045] The calculation formulas for the mean μ and standard deviation σ in step 1 are as follows:
[0046]
[0047] Among them, Yns(i) is the effective value sequence of the vibration noise sliding window, μ is the mean, and σ is the standard deviation.
[0048] As an implementation mode, in order to better illustrate the actual engineering application of the present invention, the engineering application of the present invention is further described in detail below. Figures 1 to 5 shown.
[0049] The ignition phase of the internal combustion engine is at a specific crankshaft angle, which is determined by the design. Therefore, the vibration signal is intercepted in full cycle according to the internal combustion engine working cycle, and the vibration signal sequence is converted to the crankshaft angle. If no ignition shock occurs in the ignition phase (ignition crankshaft angle), it is diagnosed as a misfire fault.
[0050] The vibration signal of a cylinder head of the same diesel engine, Figure 4 Shows the normal cylinder head vibration signal of a diesel engine. Figure 5 shows a normal shock waveform captured by the present invention, Figure 6 The vibration signal of diesel engine misfire fault is shown. Figure 7 The impact waveform captured by the present invention under the misfire fault is shown, wherein Figure 6 and Figure 7 No ignition angle impact was captured, so it was diagnosed as a misfire fault. Figure 4 and Figure 5 The box is supplemented with ignition shock, Figure 6 and Figure 7 The framed part indicates no ignition shock.
[0051] like Figure 5 As shown, based on the technical means of the present invention, Figure 4 Five sub-impact signals are intercepted from the vibration signal in FIG. , which indicates that the present invention can adaptively intercept the impact component in the vibration signal.
[0052] like Figure 7 As shown, based on the technical means of the present invention, Figure 6 Five sub-impact signals are intercepted from the vibration signal in FIG. , which once again illustrates that the present invention can adaptively intercept the impact component in the vibration signal. Figure 7 and Figure 5 Based on the impact component intercepted by the present invention, it can be known that Figure 5 There is a clear impact component in the ignition phase, and Figure 7 If there is no ignition shock component in the ignition phase, it can be judged that Figure 7 The signal shown indicates that a misfire has occurred, which indicates that the present invention can effectively diagnose a misfire in an internal combustion engine.
[0053] The vibration signal adaptive impact interception method provided by the present invention comprises the following steps:
[0054] Step 1: Calculate the sliding window vibration effective value Yns(i) of the vibration signal noise component ns(i), and calculate the mean μ and standard deviation σ of the sliding window effective value Yns(i) of the noise component.
[0055] Step 2: Use the sliding window method to calculate the sliding window effective value Y(i) of the original vibration signal s(i).
[0056] Figure 1 shows the diesel engine cylinder head vibration acceleration signal, Figure 2 shows the sliding window effective value signal calculated with window width, where N wl =17,N wl (N wl =2N w +1) is the window width, N w According to experience. Figure 2 It is known that Figure 1 After the original vibration signal in step 2 of the present invention is processed, the randomness of the original vibration signal is reduced and the impact component is highlighted.
[0057] Step 3: Take the absolute value of the vibration signal and find the peak value F1 (if there are the same peak points, take the previous sequence point) and its position P after taking the absolute value of the vibration signal. F1 , compared with the vibration signal Y(i) after the sliding window effective value calculation is completed, at the peak position PF1 Find the closest peak position P in sequence F1 The data point position is greater than the noise effective value mean μ plus three times the standard deviation σ (μ+3σ, 3σ criterion) and Then at the data point location The data between is considered as the intercepted impulse signal C1, and The signal of the segment is reset to zero.
[0058] Step 4: Repeat the above method for the zeroed vibration signal to find the peak F2 and its position P F2 , peak position P F2 Before and after data point location and Then at the data point location The data between is considered as the intercepted impact signal C2, and then The signal of the segment is reset to zero; repeat the above method to intercept The impulse signal Cn of the interval segment until the peak value F n+1 It ends when it is less than or equal to the mean of the effective value of the noise plus three times the standard deviation.
[0059] The vibration signal adaptive impact interception method provided by the present invention can realize adaptive interception of the complete vibration impact signal. The method first calculates the effective value of the noise part of the vibration signal and the sliding window of the vibration signal, as well as the mean and variance of the noise effective value; then calculates the absolute value peak of the vibration signal and its position; then finds the two points closest to the peak position whose effective values before and after the peak position are greater than the mean of the noise effective value plus 3 times the standard deviation, and the signal between the two points is the impact; in the original waveform, the impact between the two points is set to zero, and the above interception method is repeated until the peak value is less than the mean of the noise effective value plus 3 times the standard deviation.
[0060] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A vibration signal adaptive impact interception method, characterized in that: include: Calculate the mean and variance of the effective value of the noise, and obtain the absolute value peak of the vibration signal and its position; Get the two points closest to the peak position whose effective values before and after the peak position are greater than the mean of the effective value of the noise plus 3 times the standard deviation, and set the impulse between these two points to zero in the original waveform; Repeat the above interception method until the peak value is less than the mean of the noise effective value plus 3 times the standard deviation.
2. The vibration signal adaptive impact interception method according to claim 1, characterized in that: The method specifically includes: Step 1: Calculate the sliding window vibration effective value of the noise component of the vibration signal, and calculate the mean μ and standard deviation σ of the sliding window effective value of the noise component; Step 2: Use the sliding window method to calculate the sliding window effective value of the original vibration signal; Step 3: Take the absolute value of the vibration signal, find the peak value and its position after taking the absolute value of the vibration signal, and compare it with the vibration signal after the sliding window effective value calculation is completed. Find the data point position closest to the peak position and greater than the noise effective value mean plus three times the standard deviation before and after the peak position. and Obtain the intercepted impact signal C1 and The signal of the section is reset to zero; Step 4: Repeat the above method for the zeroed vibration signal to find the peak value and the data point positions before and after the peak value. and Get the intercepted impact signal C2, and then The signal of the segment is reset to zero; repeat the above steps to intercept The impulse signal Cn of the interval segment is adjusted until the peak value is less than or equal to the mean value of the noise effective value plus three times the standard deviation.
3. The vibration signal adaptive impact interception method according to claim 2, characterized in that: In step 1 and step 2, the calculation formula of the effective value of the sliding window is as follows: Where s(k) is the data point at position k in the vibration signal sequence, N wl (N wl =2N w +1) is the window width, N w According to empirical settings, Y(i) is the effective value of the window at position i in the vibration signal sequence, and n is the length of the vibration signal sequence.
4. The vibration signal adaptive impact interception method according to claim 2, characterized in that: In step 1, the calculation formula of the mean μ is as follows: Where Yns(i) is the effective value sequence of the vibration noise sliding window.
5. The vibration signal adaptive impact interception method according to claim 2, characterized in that: In step 1, the standard deviation σ is calculated as follows: Where Yns(i) is the effective value sequence of the vibration noise sliding window.
6. The vibration signal adaptive impact interception method according to claim 2, characterized in that: In step 3, the intercepted impulse signal C1 is the data point position The data between.
7. The vibration signal adaptive impact interception method according to claim 2, characterized in that: In step 4, the intercepted impulse signal C2 is at the data point position The data between.
Citation Information
Patent Citations
Reciprocating mechanical vibration signal impact starting point self-adaptive extraction method based on energy operator k-gradient
CN109598255A
Whole-period keyless phase monitoring method based on vibration mechanism and deep learning technology
CN111104887A
Method and device for intercepting vibration signal
CN113959555A
Cited By
Method and system for automatically intercepting high-frequency signal in constant-speed operation section of mechanical equipment
CN119622312A
