CN107966287A - A kind of adaptive dynamoelectric equipment Weak fault feature extracting method - Google Patents

Abstract

The present invention discloses a kind of adaptive dynamoelectric equipment Weak fault feature extracting method, gathers the vibration signal of dynamoelectric equipment first；Then two-scale matrix sequence is carried out to the initial m ultiwavelet of selection, improves the degree of approximation of m ultiwavelet scaling function；Then design meets symmetrically and translates the m ultiwavelet lifting matrixes of constraints, and then realizes that m ultiwavelet integrates building method；It is finally based on vibration signal to optimize free parameter using genetic algorithm, obtains the adaptive multi-wavelet bases function to match with fault signature, advantageous methods are provided for Weak fault feature extraction.The present invention solves the Dynamic Matching bottleneck in Weak fault characteristic extraction procedure in electromechanical equipment actual monitoring diagnostic application.

Description

An Adaptive Feature Extraction Method for Weak Faults of Electromechanical Equipment

technical field

The invention relates to the technical field of fault detection of electromechanical equipment, in particular to a method for extracting weak fault features of adaptive electromechanical equipment.

Background technique

Bearings, gears and rotors are the most common and important components of electromechanical equipment, and the failure of any one of their components will most likely lead to major economic losses and serious safety accidents. In order to avoid accidents, signal processing methods based on vibration signals are widely used in health monitoring and fault diagnosis of electromechanical equipment, such as spectrum analysis and envelope spectrum analysis. However, in the early stage of the fault, the signal features are very weak and non-stationary, which is submerged by multiple interference sources and strong noise in the electromechanical system, and the signal-to-noise ratio is low, which brings great difficulties to feature extraction and fault diagnosis.

According to the characteristics of non-stationary signals, many signal processing techniques have emerged, such as wavelet transform, sparse decomposition, empirical mode decomposition, spectral kurtosis and so on. The wavelet transform is widely used in the feature extraction of non-stationary signals by virtue of its excellent time-frequency localization characteristics and rich basis function selection, which provides a powerful tool for the non-stationary description of dynamic signals and the extraction of fault information of mechanical parts. tool. According to the study of wavelet theory, the symmetry, orthogonality, compact support, and high-order vanishing moments of wavelet basis functions are very important properties in signal processing. However, it has been proved that a single wavelet (except Haar wavelet) cannot have these properties at the same time from the mathematical aspect, which affects the engineering application effect of wavelet and limits its practical application range. As a further development of wavelet theory, multi-wavelet has excellent characteristics such as orthogonality, symmetry, compact support and high-order vanishing moment, which are very important in signal processing, and overcomes the defects and shortcomings of single wavelet. The scale function and wavelet function with different time-frequency structures provide the premise for the accuracy, reliability and comprehensiveness of mechanical fault feature extraction, making multi-wavelet quite advantageous in early fault and multiple fault diagnosis.

The essence of multi-wavelet fault feature extraction is the same as that of wavelet, which is to search for the most similar or most relevant components to the "basis function" in the signal, and the dynamic response signals caused by different types of faults during the operation of the equipment have different characteristic waveforms. . Therefore, it will be difficult to achieve the best extraction of fault characteristic signals when a fixed wavelet basis function is used for matching. However, if inappropriate and inappropriate basis functions are used for decomposition, the fault feature information will be diluted, and it will cause difficulties in fault feature extraction and diagnosis. , so that the multi-wavelet basis function has the ability to adapt to dynamic signals.

Contents of the invention

The invention aims to solve the problem of extracting weak fault features of electromechanical equipment, and provides an adaptive method for extracting weak fault features of electromechanical equipment.

In order to solve the above problems, the present invention is achieved through the following technical solutions:

A method for extracting weak fault features of adaptive electromechanical equipment, comprising the following steps:

Step 1. Collect the vibration signals that need to be monitored by electromechanical equipment;

Step 2. Construct a two-scale similar transformation matrix satisfying the conditions based on the two-scale similarity transformation theory, and use the constructed two-scale similarity transformation matrix to perform two-scale similarity transformation on the selected initial multi-wavelet to obtain the intermediate multi-wavelet;

Step 3. Based on the intermediate multi-wavelet, according to the given symmetry and translation constraints, construct the linear equation system in the symmetric lifting process, and obtain the lifting coefficient by solving the linear equation system; bring the lifting coefficient into the lifting coefficient equation, And the lifting matrix can be obtained by transformation; based on the lifting matrix, the integrated multi-wavelet is obtained according to the lifting algorithm;

Step 4. Taking the kurtosis index of the signal as the objective function, the genetic algorithm is used to solve the multi-wavelet free parameter that maximizes the objective function. This free parameter is introduced in the construction of the two-scale similar transformation matrix and the solution of the symmetric lifting coefficient equation to optimize the Adaptive multi-wavelet basis function matching the vibration signal;

Step 5. Redundant decomposition of adaptive multi-wavelet basis functions is performed to realize the translation invariance of the decomposition results, obtain more intuitive periodic pulse fault information from the time domain, realize the extraction of weak fault features, and restore the physical characteristics of faults. characteristic.

In the above step 1, the vibration signal is collected by the vibration acceleration sensor.

Compared with the prior art, the present invention uses a multi-wavelet integration construction method to obtain a large difference multi-wavelet basis function with symmetry and parameter regulation, and realizes the best matching between the wavelet basis function and the fault characteristic waveform through self-adaptive optimization, which is beneficial to The best extraction of fault features, the present invention has the following significant advantages:

1) The present invention constructs an integrated multi-wavelet basis function with symmetric or antisymmetric waveforms, which ensures the linear phase of the filter, avoids phase distortion during decomposition and reconstruction, and can obtain the analysis result of zero phase shift in the analysis process . At the same time, it is beneficial to the boundary processing in the operation process and reduces boundary distortion. Therefore, it has significant advantages in the process of fault feature extraction from the time domain;

2) The present invention uses the kurtosis index as the optimization criterion in the adaptive matching process, and the kurtosis index has a significant advantage on the early fault impact characteristics, and uses the genetic algorithm as an optimization method to avoid the mathematical expression between the objective function and the free variable , making full use of the robustness of genetic computing and the advantages of global parallel search, and obtaining multi-wavelet basis functions suitable for early fault features, which have corresponding advantages in early fault feature extraction;

3) The present invention can increase the approximation order and vanishing moment to any order according to the actual engineering requirements through the two-scale similarity transformation and the symmetric lifting framework, greatly improving the properties of the integrated multi-wavelet basis functions, and obtaining symmetric or antisymmetric basis functions. Provide possibility for weak fault feature extraction;

4) The present invention can be used for fault feature extraction and fault diagnosis of large-scale electromechanical equipment based on vibration monitoring, so as to avoid sudden accidents and reduce economic losses.

Description of drawings

Figure 1 is a flowchart of a method for extracting weak fault features of adaptive electromechanical equipment.

Figure 2 is the waveform diagram of the original signal of the faulty bearing.

Figure 3 is a spectrum diagram of a faulty bearing.

Figure 4 is the envelope spectrum diagram of the faulty bearing.

Fig. 5 is a diagram of two basis functions after adaptation: (a) multi-wavelet wavelet function ψ ₁ , (b) multi-wavelet wavelet function ψ ₂ .

Figure 6 shows the results of adaptive multi-wavelet redundancy decomposition: (a) the detail signal corresponding to ψ ₁ , and (b) the detail signal corresponding to ψ ₂ .

Detailed ways

The present invention is proposed based on the two-scale similarity transformation of multi-wavelet and the symmetric lifting frame, which first collects the vibration signal during the operation of the mechanical equipment through the vibration acceleration sensor; then performs two-scale similarity transformation on the selected initial multi-wavelet to improve the multi-wavelet scale The approximation order of the function; then design a multi-wavelet lifting matrix that satisfies the symmetry condition to ensure the linear phase characteristics of the filter of the multi-wavelet basis function, avoid phase distortion during signal decomposition and reconstruction and improve boundary processing capabilities; by studying the similarity between the two scales Transformation and symmetric lifting frame multi-wavelet integration construction algorithm, design a multi-parameterized two-scale similar transformation control matrix and multi-wavelet symmetric lifting matrix that meet the constraints, and realize the multi-wavelet integration construction method; finally, based on the vibration signal, use the genetic algorithm to carry out the free parameters Optimizing to obtain an adaptive multi-wavelet basis function matching the fault features, which provides a favorable means for the extraction of weak fault features. The invention solves the dynamic matching bottleneck in the weak fault feature extraction process in the actual monitoring and diagnosis application of electromechanical equipment, and proposes a novel multi-degree-of-freedom self-adaptive signal processing method.

Specifically, a method for extracting weak fault features of adaptive electromechanical equipment includes the following steps:

The first step: collect the monitoring vibration signal of the electromechanical equipment system, and provide the original information source for the subsequent selection of control parameters through optimization methods. For key rotating parts such as bearings, gears and rotors in electromechanical systems, vibration sensors are used to obtain corresponding vibration signals.

Step 2: Perform two-scale similarity transformation on the initial multi-wavelet. Based on the two-scale similarity transformation theory, a parameterized matrix that satisfies specific conditions (such as symmetry, large difference, reversibility, etc.) is designed to obtain a multi-wavelet basis function library with greater degrees of freedom, and to improve the multi-wavelet properties by increasing the approximation order.

Let H(ω) and G(ω) be the low-pass and high-pass filter symbols of multi-wavelet multi-scale functions, H ⁽ⁿ⁾ (ω) approximation order n, G ^(m) (ω) disappearance order m, and Spectral radius ρ(H ⁽ⁿ⁾ (0))<N, N is a positive integer greater than 1. Further, it is assumed that H ⁽ⁿ⁾ (0) has an eigenvalue of 1, corresponding to the right eigenvector r _n , such that H ⁽ⁿ⁾ (0)r _n =r _n .

(1) Select a two-scale matrix M(ω) that satisfies the conditions, and add symmetry constraints at the same time, so that the two-scale similarity matrix M(ω) satisfies: where E(ω), Both are diagonal matrices, is the symmetry point of the multi-scale function, T _j is the symmetry point of the multi-wavelet function, and diag(·) represents the diagonal matrix.

(2) Structure

(3) Find a suitable r _n+1 corresponding to H ⁽ⁿ⁺¹⁾ (0) with an eigenvalue of 1;

(4) Repeat the first three steps until H(ω) of the required approximation order is obtained.

In the above algorithm, only the initial H ⁽ⁿ⁾ (ω) and M(ω) of each cycle need to be selected. After each cycle, the approximation order and regularity of the multi-scale function corresponding to H ⁽ⁿ⁾ (ω) are increased by one order, but unfortunately the vanishing moment of the multi-wavelet basis function corresponding to G ^(m) (ω) Down one notch. If M(ω) is a trigonometric polynomial, and |M(ω)| is linear at z=e ^-iω , then the H ⁽ⁿ⁺¹⁾ (ω) constructed by the algorithm is a trigonometric polynomial and maintains a multi-scale function tightness and symmetry.

The multiwavelet transformed by the above-mentioned two-scale similarity is called the intermediate multiwavelet {Φ _p , Ψ _p }, and the symbols of the corresponding low-pass and high-pass filters are H _p (ω) and G _p (ω), respectively. The approximation order of the original multiwavelet and the intermediate multiwavelet are improved, its regularity is also enhanced, and the properties of the multiscale function are also improved. However, at the cost of reducing the vanishing moment of the multiwavelet basis function Ψ _p , it weakens the smoothness and local positioning ability of the multiwavelet function, which has an adverse effect on the signal processing process.

Step 3: Perform lifting frame transformation on the intermediate multi-wavelet. Based on the symmetrical lifting framework, symmetry constraints and translation constraints are added, and according to the lifting coefficient equation, a multi-wavelet parameterized lifting matrix is designed to ensure the linear phase characteristics of the multi-wavelet basis function filter.

The n-order continuous moments of intermediate multi-wavelet scaling functions and wavelet functions {Φ _p ,Ψ _p } are M(Φ _p ,,n)=∫Φ _p (x)x ⁿ dx and M(Ψ _p ,n)=∫Ψ _p (x)x ⁿ dx. then there is

With the help of the moment calculation formula, the multi-wavelet construction is used for calculation. The process of using the lifting method to construct multi-wavelets can be expressed as: first select the initial multi-wavelet ω ₀ (x), where ω ₀ (x)=ψ ₁ or ψ ₂ , and then select other basis functions ω ₁ for modifying the multi-wavelet (x),...,ω _k (x) translation k, and finally can construct a new multi-wavelet through the "lifting coefficient equation" The lift coefficient equation is:

Compared with single wavelet, multi-wavelet has more advantages in lifting construction. For example, in single wavelet lifting, the scale function can only be used to correct the original wavelet function, while in multi-wavelet lifting, it is used to correct a certain multi-wavelet function. not only includes two multiscale functions, but also another corresponding multiwavelet function, that is, for ψ ₁ , for Obviously, there are more basic functions used to construct new multi-wavelet functions than single wavelets, which brings greater freedom and flexibility to the construction of multi-wavelets to meet more and more specific requirements.

Assuming that the vanishing moment of the multi-wavelet is raised from p to p', and integrating both sides of the "lifting coefficient equation", the following lifting linear equations can be obtained:

The integral value in the above formula is calculated by using the moment calculation formula, and the solution { _ci} of the equation system is the coefficient of the multi-wavelet lifting function. A multiwavelet lifting framework can be obtained by Z-transforming the lifting coefficient equation.

The above-mentioned multi-wavelet lifting process cannot guarantee the symmetry of the multi-wavelet function after lifting. In order to ensure the symmetry of the lifting process, the "symmetry selection" method is used to select the translation amount of other functions used to modify the multi-wavelet. Assuming an initial multiscale function It is symmetric or antisymmetric to the multiwavelet functions ψ ₁ and ψ ₂ , and the symmetry centers are respectively Then the symmetric lifting method is expressed as follows, taking the symmetric lifting of ψ ₁ as an example, the translation of the lifting function must meet

In the formula: i=1,2; j=1,2,...; m=1,2.

make Respectively represent the symmetry properties of the initial multiscale function and the initial multiwavelet function, where 1 represents symmetry and -1 represents anti-symmetry. Will and M(ψ _i ,k,n)=∫ψ _i (x+k)x ⁿ dx are substituted into the lifting linear equation system, and the first matrix on the left side of the equal sign represents M _B , where M _B =MB, M and B respectively

Let B be a symmetric matrix

The coefficient vectors in the lifted linear equation system are expressed as And the right side of the equation is expressed as M _ψ =[M(ψ _i ,0,p),M(ψ _i ,0,p+1),…M(ψ _i ,0,p'-1)] ^T , then the formula into the following formula

M _B C = M _ψ

The solution C of the equation is the coefficient used to lift ψ ₁ , the situation of ψ ₂ is similar, the only difference is that the function of lifting ψ ₂ is and Substitute the lifting coefficient into the formula lifting coefficient equations, and perform Z transformation to obtain the lifting matrices T and S. Therefore, adaptive multi-wavelet integration can be realized by means of lifting matrices T and S, as follows:

H _s (ω) = H _p (ω)

G _s (ω)＝T(ω ² )(G _p (ω)+S(ω ² )H _p (ω))

In the formula: H _s , G _s are the symbols of low-pass filter and high-pass filter of integrated multi-wavelet, respectively.

Compared with the initial multi-wavelet, the integrated multi-wavelet {Φ _s , Ψ _s } has improved approximation order and vanishing moment of scaling function and wavelet function, enhanced regularity, smoothness and local positioning ability, and improved performance , the most important thing is that they guarantee the symmetry of multiwavelets, which is beneficial to their application in signal feature extraction, especially in time domain analysis.

The fourth step: through the two-scale similarity transformation and the symmetric lifting frame integration construction algorithm, the parameterized multi-wavelet construction is realized, and the parameterized control multi-wavelet basis function with excellent properties such as linear phase is obtained.

Free parameters will be introduced in the construction of the two-scale similar transformation matrix and the solution of the symmetric lifting coefficient equation. These non-zero free parameters are the key to achieve adaptive construction. In view of the sensitivity of the kurtosis index to the early shock signal, the present invention takes the kurtosis index of the detail signal as the objective function, and solves the multi-wavelet free parameter that maximizes the objective function K _P to optimize the adaptive multi-wavelet wavelet that matches the early fault characteristics. Wavelet basis functions.

Objective function K _P definition:

In the formula: x—detail signal; p(x)—probability density of signal amplitude.

In view of the robustness of the genetic algorithm and the advantages of global and parallel search, it does not require the mathematical expression between the objective function and variables. The invention adopts the genetic algorithm and uses the objective function K _P as the fitness function to construct the multi-wavelet base function adaptively matching the signal characteristics, and completes the optimal construction of the self-adaptive multi-wavelet.

The adaptive matching of adaptive multi-wavelet ensemble construction is completed based on genetic optimization algorithm and kurtosis maximization optimization criterion.

Step 5: Use the adaptive optimal multi-wavelet basis function to perform redundant decomposition, realize the translation invariance of the decomposition results, obtain more intuitive periodic pulse fault information from the time domain, and realize the extraction of weak fault features , to restore the physical properties of the fault.

The present invention combines the advantages of the multi-wavelet two-scale similarity transformation and the symmetric lifting framework, and constructs a multi-wavelet multi-scale function with high approximation order and strong regularity, so that the energy of the signal in the frequency domain is more concentrated; through the symmetric lifting process, the The vanishing moment of the multi-wavelet function is improved, the localization ability of the basis function is improved, the linear phase or generalized linear phase of the filter is also guaranteed, the error caused by decomposition and reconstruction is avoided, and the boundary processing ability is improved, so that more accurate Describe and express higher-order complex signals. By designing the free parameters in the construction process and selecting specific basis function evaluation and optimization criteria, combined with genetic algorithm and other optimization methods to optimally select the free parameters in the construction process, and then realizing the adaptive multi-wavelet basis function construction for the signal to be analyzed, In order to realize the extraction of weak fault features.

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific examples.

Rolling bearings are one of the most common and most important key components in electromechanical equipment. However, working for a long time in complex operating environments and operating conditions will lead to inevitable damage to rolling bearings, resulting in paralysis of electromechanical equipment, resulting in major economic losses or personnel. casualties. In order to prevent and avoid major accidents, it is necessary to discover the faults and hidden dangers in the process of equipment operation as early as possible, and eliminate the faults in the bud. In this embodiment, the GHM is used as the initial multi-wavelet, and the self-adaptive integrated multi-wavelet structure is used to analyze the vibration signal of the bearing monitoring in the mechanical equipment, and to extract the fault characteristics of the rollers in the rolling bearing from the time domain.

This bearing is based on the integrated multi-wavelet construction method and redundancy decomposition embodiment of vibration signal feature extraction. An adaptive electromechanical equipment weak fault feature extraction method is designed, as shown in Figure 1. It mainly includes the following steps:

The first step: use the vibration acceleration sensor to collect the vibration signal of the bearing.

The acquisition system includes YE6267 dynamic data collector and CA-YD-117 piezoelectric acceleration sensor of Jiangsu Lianeng Electronic Technology Co., Ltd. The performance index of the acceleration sensor is shown in Table 1. The collector realizes 16-bit A/D parallel data acquisition based on the USB2.0 interface, and the signal acquisition and monitoring is completed by the lower computer (monitoring front-end computer).

Table 1 CA-YD-117 piezoelectric acceleration sensor characteristic parameter list

The bearing model used is 552732QT cylindrical roller bearing, which is installed on the input shaft of the gearbox. The bearing structure parameters are listed in Table 2. In this example, a slight fault on the outer ring of a rolling bearing is used to verify the effectiveness of the method. The damage size in the test is a slight scratch fault on the outer ring of 1*1mm.

Table 2 552732QT Bearing Structure Parameters

The fault characteristic frequency of the rolling bearing outer ring can be calculated by the following formula:

In the formula: z—number of rolling elements; f _r —rotation frequency/Hz; d—diameter of rolling elements/mm; D—outer diameter of bearing/mm; α——contact angle/°.

During the test, the vibration acceleration sensor was installed at the faulty bearing seat, the input shaft speed was 673r/min, the sampling frequency was 12.8kHz, and the sampling length was 4096 points. The root and rolling bearing outer ring fault formula (1) obtained the bearing outer ring fault characteristic frequency 82.8Hz, that is, the time interval is 12.7ms.

Figure 2 shows the vibration time-domain signal of the equipment when the outer ring of the bearing is faulty, Figure 3 is the frequency spectrum of the faulty bearing signal, and Figure 4 is the Hilbert envelope spectrum of the faulty bearing signal, from which we can see the weak impact signal generated by the fault Submerged in the strong background noise, it is difficult to find the fault characteristics of periodic pulses, and the spectral lines corresponding to the fault characteristic frequency in the frequency domain are also covered in the noise spectral lines, making it difficult to judge whether the bearing has a fault.

The second step: GHM is the most common and one of the most widely used dual multi-wavelet basis functions in engineering. It has compact support, orthogonality, symmetry, second-order approximation order and vanishing moment. Here, the GHM multi-wavelet is used as the initial multi-wavelet. The symmetry point or anti-symmetry point of the GHM multi-wavelet multi-scale function and the multi-wavelet basis function is 1/2 or 1, and at the same time, H(0) has an eigenvector corresponds to an eigenvalue of 1. Therefore, according to the conditions that must be satisfied in constructing the two-scale similarity transformation matrix, combined with the characteristics of GHM multi-wavelet, the two-scale similarity transformation matrix M(ω) based on GHM multi-wavelet is constructed as follows:

In the formula: a, b——non-zero parameters.

Let H _G (ω) and G _G (ω) be the low-pass and high-pass filter symbols of the GHM multi-wavelet scaling function and multi-wavelet function, according to the similarity transformation of the two scales:

After the above transformation, a new intermediate function {Φ _p , Ψ _p } is obtained. The multi-scale function Φ _p has a third-order approximation order, while the vanishing moment of the multi-wavelet basis function Ψ _p is reduced to the first order, which weakens the multi-wavelet The smoothness and local localization ability of the function.

Step 3: Based on the intermediate multiwavelet {Φ _p , Ψ _p }, assuming that the vanishing moment is raised from order 1 to order 5, set B _ωi =±1(ω _i =ψ _i or 1 represents symmetry, -1 represents anti-symmetry) represents the symmetry or anti-symmetry of the initial multi-wavelet, then the linear equation in the process of symmetry promotion:

The solution of equation (4) is the lift coefficient of Ψ ₁ , and the lift of Ψ ₂ is also similar to Ψ ₁ . After obtaining the lifting coefficient, bring it into the lifting coefficient equation, and get the lifting matrix T and S through Z transformation. Based on the lifting matrices T and S, the new multi-wavelet after integrated construction is obtained according to the following lifting algorithm:

The multi-wavelet {Φ _s , Ψ _s } constructed through the lifting frame has 3rd order of approximation and 5th order of vanishing moment.

Step 4: In the previous construction process, the two-scale similarity transformation introduces two free parameters, and when the linear equation M _B C = M _ψ of the symmetric lifting process is an underdetermined linear equation system, there is N _f = 4 -Rank(M _B ) free parameters. These free parameters will affect the waveform of the multiwavelet basis function. However, the kurtosis index is very sensitive to the shock response characteristics caused by early local faults. Therefore, the adaptive construction process of the above-mentioned integrated multi-wavelet takes the kurtosis index of the detail signal as the objective function, and _solves the multi-wavelet free parameters to optimize adaptive multiwavelet basis functions that match a given signal.

Objective function:

In the formula: x—detail signal; p(x)—probability density of signal amplitude.

It can be seen from the expression of the objective function K _P that there is no direct connection between the principle of maximization of kurtosis and these free parameters, and the relationship between them is very complicated. Common optimization algorithms often require a complete relational expression, which is difficult to implement here . In view of the strong robustness of the genetic algorithm and the characteristics of global and parallel search, it does not need the mathematical expression between the objective function and the variables. In this section, the genetic algorithm is used, and the objective function K _P is used as the fitness function to construct a multi-wavelet adaptive matching signal feature. The range of free parameters is selected as [-50,0)U(0,50], and the arithmetic crossover algorithm Sub and non-uniform mutation operators, the population size is 100, the initial population number is set to 50, the crossover probability is set to 0.6, and the mutation probability is set to 0.05.

After the optimization of the above process, the multi-wavelet basis functions constructed by adaptive multi-wavelet integration are shown in Figure 5.

Step 5: After the integrated multi-wavelet adaptive construction is completed, the multi-wavelet redundant decomposition is performed on the signal to realize the translation invariance of the decomposition results. The specific implementation process is as follows:

The decomposition process of redundant multiwavelet transform is the result of matrix interpolation and zero padding of the previous decomposition layer by the low-pass and high-pass filters of the subsequent decomposition layer, and matrix interpolation zero padding refers to the insertion of zeros with a size of r×r matrix. Let T represent the interpolation zero padding operator, then for any integer i,

(Tx) _2i ＝x _i , (Tx) _2i+1 ＝0

Then the low-pass and high-pass filter coefficients {H, G} are calculated by the following equation when the decomposition layer is l in the process of redundant multiwavelet transform:

(1) When k is not an integer multiple of 2 ^l ,

(2) In other cases,

The use of redundant multiwavelet transform is beneficial to improve the accuracy of feature extraction and spectral accuracy of decomposition results, and then achieve more accurate feature extraction and fault diagnosis of mechanical equipment faults.

After three layers of redundancy decomposition, the branch with the largest kurtosis is selected, and the result is shown in Figure 6. It can be seen from Figure 6 that this method can successfully extract the periodic pulse signal of the outer ring fault with a period of 12.7 ms, so it can be seen that this method has an obvious effect on enhancing the weak fault characteristics, and can extract the fault from the strong background Extract the fault features of rolling bearing outer ring from the noise.

The present invention proposes an integrated construction method of self-adaptive multi-wavelet basis functions on the basis of in-depth study of two-scale similarity transformation and symmetric lifting framework, and adopts the kurtosis index most sensitive to the impact pulse of early fault as the self-adaptive process The optimization criterion is optimized by genetic algorithm, the vibration signal collected by the sensor is analyzed, and the adaptive integrated multi-wavelet basis function with excellent properties matching the fault characteristics is obtained, so as to realize the extraction of weak fault features. The present invention improves the approximation order of the multi-scale function and the vanishing moment of the multi-wavelet basis function, simultaneously constrains the symmetry of the basis function, ensures the linear phase and the generalized linear phase of the filter, improves the performance of the original multi-wavelet, and avoids decomposition The phase distortion that appears in the process improves the boundary processing ability, constructs an adaptive multi-wavelet basis function that is suitable for dynamic signals, and maintains the excellent characteristics of the initial multi-wavelet such as tight support and symmetry, which is an important feature of electromechanical equipment. Weak fault feature extraction provides a new method.

It should be noted that although the above-mentioned embodiments of the present invention are illustrative, they are not intended to limit the present invention, so the present invention is not limited to the above specific implementation manners. Without departing from the principles of the present invention, all other implementations obtained by those skilled in the art under the inspiration of the present invention are deemed to be within the protection of the present invention.

Claims (2)

1. a kind of adaptive dynamoelectric equipment Weak fault feature extracting method, it is characterized in that, including step is as follows：

The vibration signal of monitoring needed for step 1, collection dynamoelectric equipment；

Step 2, the two-scale matrix sequence matrix for meeting based on two-scale matrix sequence theory building condition, and using being constructed Two-scale matrix sequence matrix two-scale matrix sequence is carried out to selected initial m ultiwavelet, obtain middle m ultiwavelet；

Step 3, based on middle m ultiwavelet, symmetrical lifting process is constructed according to given symmetry and translational movement constraints In system of linear equations, and got a promotion coefficient by solving system of linear equations；Bring Lifting Coefficients into Lifting Coefficients equation, and By converting the matrix that can get a promotion；Based on lifting matrixes, integrated m ultiwavelet is obtained according to boosting algorithm；

Step 4, using the kurtosis index of signal as object function, using genetic algorithm solve make object function maximum m ultiwavelet oneself By parameter, which introduces in two-scale matrix sequence matrix construction and the symmetrical Lifting Coefficients equation of solution, with preferred Go out the adaptive multi-wavelet bases function to match with vibration signal；

Step 5, carry out redundancy resolution to adaptive multi-wavelet bases function, the translation invariance of decomposition result is realized, from time domain More intuitively periodically pulsing fault message is obtained, the extraction of Weak fault feature is realized, reduces the physical characteristic of failure.

2. a kind of adaptive dynamoelectric equipment Weak fault feature extracting method according to claim 1, it is characterized in that, step In 1, vibration signal is gathered by vibration acceleration sensor.