CN111413404A - On-line measurement method of blade crack based on blade tip timing and support vector machine principle - Google Patents

Abstract

The invention relates to an on-line measurement method for blade cracks based on blade tip timing and support vector machine principles. Multiple blade tip timing sensors are arranged at different positions of a casing of a rotating machine, and under simulated working conditions, blades are measured through the blade tip timing sensors. The arrival time of the timing sensor at the tip of each blade is measured, and the vibration displacement, amplitude and vibration frequency of the blade are measured. The vibration data of the blade, including the vibration data of the blade without cracks and the blades with cracks, are marked as positive samples. , the vibration data of cracked blades are marked as negative samples, which are sent to the support vector machine for classification training and a classification model is established.

Description

On-line measurement method of blade crack based on blade tip timing and support vector machine principle

technical field

The invention belongs to the field of rotating machinery state monitoring, in particular to an on-line measurement method for blade cracks based on blade tip timing and support vector machine principles.

technical background

Large-scale rotating machinery includes large-scale equipment such as aero-engines and steam turbines, and is the core component of various military and commercial aircraft in the aerospace field, as well as key equipment such as generator sets and steam generators in the industrial field. In particular, blades are the core components of rotating machinery, and their working states directly affect the work efficiency and safe and stable operation of these key equipment. The working environment of rotating machinery blades is very harsh, and they are exposed to severe conditions such as high stress, high and low temperature or high erosion for a long time. These external conditions will produce complex periodic or non-periodic stress on the blades of large rotating machinery. When the stress exceeds the yield strength limit of the blade material, cracks will occur in the blade, and then the blade may break. Blade cracks are one of the main causes of large-scale rotating machinery failures. Therefore, accurate measurement of the vibration parameters of the blades and online measurement of whether the blades have cracks can provide real-time early warning of blade failures. Condition monitoring and fault diagnosis have very important practical significance.

The rotating blade vibration measurement technology based on the blade tip timing principle ^[1-3 ] is a typical non-contact measurement method. The on-line measurement of blade vibration parameters such as vibration displacement, amplitude and vibration frequency is realized by using relevant mathematical algorithms. Compared with traditional off-line blade condition detection methods and online detection methods such as strain gage method, frequency modulation method and acoustic method, blade tip timing technology has the advantages of non-contact, real-time online measurement of all blades, etc. practicality. Support Vector Machine is a machine learning method based on statistical theory, which can be used for classification problems of linear and nonlinear data. The support vector machine was originally designed to solve the binary classification problem. Classify the tested samples. Therefore, after the vibration data of normal blades and cracked blades are obtained by the blade tip timing method, the support vector machine algorithm is used to train and establish a classification model, so as to realize the online measurement of blade cracks under the actual operating conditions of rotating machinery.

At present, the crack measurement of rotating machinery blades relies on offline detection technology, which cannot meet the actual needs of online measurement of blade cracks in large rotating machinery.

[1] Ouyang Tao. Rotating blade vibration detection and parameter identification technology based on blade tip timing [D]. Tianjin University, 2011.

[2] Zhao Xingming, Teng Guangrong, et al. New technology of blade tip timing rotating blade vibration measurement [J]. Measurement and Control Technology, 2006(03):17-19.

[3] Wang Ping. Review of the application of blade tip timing method in vibration measurement of foreign aero-engine blades [J]. Aeronautical Science and Technology, 2013(06):9-13.

SUMMARY OF THE INVENTION

The purpose of the present invention is to aim at the deficiencies of the above-mentioned prior art, provide a kind of blade crack on-line measurement method based on blade tip timing and support vector machine principle, obtain the vibration data of the crackless blade and the cracked blade by the blade tip timing method, The support vector machine algorithm is used to classify and train the blades with and without cracks, and a support vector machine classification model is established to realize the online measurement of cracks in rotating machinery blades under working conditions.

The technical scheme of the present invention is as follows:

An online measurement method for blade cracks based on blade tip timing and support vector machine principles. Multiple blade tip timing sensors are set at different positions of the casing of the rotating machine. Under simulated working conditions, the blade tip timing sensor measures the arrival of each blade The arrival time of the blade tip timing sensor, the vibration displacement, amplitude and vibration frequency of the blade are measured. The blade vibration data includes the vibration data of the non-cracked blade and the cracked blade. The vibration data of the non-cracked blade is marked as a positive sample with cracks. The vibration data of the blade are marked as negative samples, which are sent to the support vector machine for classification training and a classification model is established. Include the following steps:

(1) Set up multiple blade tip timing sensors at different positions of the rotating machinery casing to measure the time when the blade reaches each blade tip timing sensor at different speeds;

(2) Carry out a simulated working condition test when the blade is intact and without cracks, and send the arrival time signal of each blade measured by the blade tip timing sensor to the blade condition monitoring system. Combined with the installation positions of multiple blade tip timing sensors and the difference in rotating machinery Rotation speed and blade tip timing algorithms are used to calculate the vibration displacement, amplitude, vibration frequency, initial phase, vibration constant deviation and resonance multiplier of the blade without cracks at different speeds;

(3) For the cracked blade, carry out a simulated working condition experiment to obtain the vibration displacement, amplitude, vibration frequency, initial phase, vibration constant deviation and resonance frequency multiplier of the cracked blade at different speeds;

(4) Set the vibration data of the blade without cracks as a positive sample, and set the data label to 0; set the vibration data of the blade with cracks to be a negative sample, and set the data label to 1;

(5) Taking the total number of negative samples as a reference, the positive samples are sampled by the equal interval downsampling method to achieve the balance of positive and negative samples, and the balanced positive samples and negative samples are combined into a training set;

(6) Use the min-max normalization algorithm to normalize the data in the training set to eliminate the dimension;

(7) According to the training set after eliminating the dimension, use the support vector machine algorithm to perform classification training, and establish a support vector machine classification model;

(8) When the rotating machine is running in the working state, according to the arrival time of the blade obtained by the blade tip timing sensor, the blade state monitoring system is used to calculate the vibration displacement, amplitude, vibration frequency, initial phase, Vibration constant and resonance frequency multiplier blade vibration data;

(9) The blade vibration data measured in the working state is used as the sample to be tested, and after normalization, it is sent to the SVM classification model trained in the blade state monitoring system, and the pending SVM classification model is calculated by the SVM classification model. The test sample label is 0 or 1,

Realize online measurement of cracks in rotating machinery blades.

The beneficial effects and advantages of the present invention are as follows:

The method of the invention overcomes the shortcomings of the existing rotating machinery blade crack measurement technology, and provides an online blade crack measurement method based on the blade tip timing and the support vector machine principle. And the vibration data of the cracked blade, combined with the support vector machine algorithm to establish the classification model of the crackless blade and the cracked blade, and use the trained classification model to realize the online measurement of the crack of the rotating machinery blade.

Description of drawings

Selected embodiments of the present invention are described in the following figures, which are illustrative and not exhaustive or limiting, in which:

Figure 1 shows the structure diagram of blade crack online measurement based on blade tip timing and support vector machine principle

Figure 2 shows the schematic diagram of the support vector machine classification algorithm

Description of the labels in the figure:

In Figure 1: 1 is the tip timing sensor A; 2 is the tip timing sensor B; 3 is the tip timing sensor C; 4 is the tip timing sensor D; 5 is the rotating machine casing; 6 is the blade; 7 is the blade Blade condition monitoring system;

In Figure 2: 8 is the optimal classification hyperplane; 9 is the positive class support vector; 10 is the negative class support vector.

Detailed ways

The steps of the present invention are described in detail below, and are intended to be described as embodiments of the present invention, not the only form that can be manufactured or utilized, and other embodiments that can achieve the same function should also be included within the scope of the present invention.

The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

The purpose of the present invention is to overcome the deficiency that the existing rotating machinery blade cracks cannot be measured online, and proposes an online blade crack measurement method based on the blade tip timing and the support vector machine principle;

(1) The online measurement structure of blade crack based on blade tip timing and support vector machine principle is shown in Figure 1. Four blade tip timing sensors are installed at any different positions of the rotating machinery casing 5, including the blade tip timing sensor A1, Tip timing sensor B2, tip timing sensor C3 and tip timing sensor D4, four tip timing sensors can measure the time when each blade reaches the tip timing sensor in the same circle;

(2) When the blade has no cracks, put the rotating machine into the rotating machine condition simulation test bench to simulate the operation of the rotating machine. Assuming that the rotating machine rotates clockwise, when the blade 6 is stimulated to vibrate synchronously, the four The arrival time signal of the blade 6 measured by the blade tip timing sensor is sent to the blade condition monitoring system 7. According to the installation positions of the four blade tip timing sensors, the different rotational speeds of the rotating machinery and the blade excited vibration response equation, the blades 6 in the same circle are successively The vibration displacement equations of the four blade tip timing sensors are:

为叶片6的初始相位，C为叶片6的振动常偏量，N为叶片6的共振倍频数，α₁为叶尖定时传感器B2相对于叶尖定时传感器A1的安装弧度角，α₂为叶尖定时传感器C3相对于叶尖定时传感器A1的安装弧度角，α₃为叶尖定时传感器D4相对于叶尖定时传感器A1的安装弧度角，利用遍历算法将N所有的可能取值带入式(1)，结合最小二乘法可以求解叶片6无裂纹时的振动频率ω、初始相位

振幅A和振动常偏量C；In the formula, y ₀ is the vibration displacement of the blade 6 when it passes the tip timing sensor A1, y ₁ is the vibration displacement of the blade 6 when it passes the tip timing sensor B2, and y ₂ is the vibration displacement of the blade 6 when it passes the tip timing sensor C3. , y ₃ is the vibration displacement of the blade 6 when it passes through the tip timing sensor D4, A is the amplitude of the blade 6,

is the initial phase of the blade 6, C is the vibration constant deviation of the blade 6, N is the resonance multiplier of the blade 6, α1 is the installation radian angle _of the blade tip timing sensor B2 relative to the blade tip timing sensor A1, _α2 is the blade The installation radian angle of the tip timing sensor C3 relative to the tip timing sensor A1, α3 is the installation radian angle _of the tip timing sensor D4 relative to the blade tip timing sensor A1, and the traversal algorithm is used to bring all possible values of N into the formula ( 1), combined with the least squares method, the vibration frequency ω and initial phase of the blade 6 without cracks can be solved

Amplitude A and vibration constant C;

振幅A'、振动常偏量C'和共振倍频数N'；The blade 6 is artificially manufactured with the required cracks and then loaded into the rotating machine, and the above-mentioned simulation operation process is repeated by using the rotating machine working condition simulation test bench again, and the vibration displacement y' of the cracked blade 6 passing through the tip timing sensor A1 at the same rotational speed can be obtained. _0. The vibration displacement y ₁ ' when passing the tip timing sensor B2, the vibration displacement y' ₂ when passing the tip timing sensor C3 and the vibration displacement y' ₃ when passing the tip timing sensor D4, and obtain the cracked blade 6 vibration frequency ω', initial phase

Amplitude A', vibration constant C' and resonance frequency N';

C和N设为正类样本、数据标签设置为0；将有裂纹叶片6在不同转速下测得的叶片振动数据y'₀、y₁'、y'₂、y'₃、A'、

C'和N'设为负类样本，数据标签设置为1；(3). The blade vibration data y ₀ , y ₁ , y ₂ , y ₃ , A, and

C and N are set as positive samples, and the data label is set as 0; the blade vibration data y' ₀ , y ₁ ', y' ₂ , y' ₃ , A',

C' and N' are set as negative class samples, and the data label is set to 1;

(4) Support vector machine is a machine learning method based on statistical learning theory, suitable for linear and nonlinear data classification, as shown in Figure 2, its main idea is to find an optimal classification hyperplane 8, so that the training concentration The distance between the positive class samples and the negative class samples to the optimal classification hyperplane 8 is the largest, and the optimal classification hyperplane 8 is used to classify the tested samples, and the points in the positive class samples of the test set are closest to the optimal classification hyperplane 8 It is called the positive class support vector 9, and the point closest to the optimal classification hyperplane 8 in the negative class samples of the test set is called the negative class support vector 10, and the determination of the optimal classification hyperplane 8 only depends on the positive class sample support vector 9 and negative class sample support vector 10, non-support vector points do not work;

Since the aforementioned positive samples are usually more than negative samples, the total number of negative samples is used as a reference, and a large number of positive samples are sampled by the equal interval downsampling method to achieve positive and negative sample balance. The negative class samples are combined into a training set, each row in the training set represents a sample, and each column represents the measured value of the blade vibration data;

Since the data units in the training set are different and the magnitudes of the values are also different, the maximum and minimum normalization method is used to eliminate the dimension of the training set to improve the quality of the training set. The minimum and maximum normalization equation is:

为训练集第j列归一化后的数据，x_ij为训练集第j列归一化前的数据，

为训练集第j列数据归一化前的均值，x_jmax是训练集第j列数归一化前的最大值，x_jmin是训练集第j列数据归一化前的最小值，消除量纲后的训练集送入支持向量机进行分类训练，支持向量机的核函数选择高斯径向基核函数，并利用网格搜索和交叉验证方法对高斯径向基核函数的核宽和支持向量机的软间隔惩罚系数进行参数优化，求解最优参数，建立支持向量机分类模型；In the formula,

is the normalized data of the jth column of the training set, x _ij is the data before the normalization of the jth column of the training set,

is the mean value of the data in the jth column of the training set before normalization, x _jmax is the maximum value before the normalization of the jth column of the training set, and x _jmin is the minimum value before the normalization of the jth column of the training set. The post-class training set is sent to the support vector machine for classification training. The kernel function of the support vector machine selects the Gaussian radial basis kernel function, and uses grid search and cross-validation methods to analyze the kernel width and support vector of the Gaussian radial basis kernel function. The soft interval penalty coefficient of the machine is used to optimize the parameters, solve the optimal parameters, and establish a support vector machine classification model;

C”和N”，并作为待测样本，每一个待测样本利用下式进行同比例归一化：(5) When the rotating machine is running in the working state, the blade vibration data y” ₀ , y” ₁ , y” ₂ , y” ₃ , A of the blade 6 in different rotating speed working states can be obtained by the aforementioned blade tip timing method ",

C" and N" are used as samples to be tested, and each sample to be tested is normalized in the same proportion using the following formula:

为待测样本同比例归一化后第j列的数据，X_ij为待测样本同比例归一化前第j列的数据；In the formula,

is the data of the jth column after normalization of the sample to be tested in the same proportion, X _ij is the data of the jth column before the normalization of the same proportion of the sample to be tested;

(6) Send the sample to be tested normalized to the same proportion into the trained SVM classification model. If the SVM classification model calculates that the data label of the sample to be tested is 0, it means that the blade 6 has no cracks. , if the support vector machine classification model calculates that the data label of the sample to be tested is 1, it means that a crack has occurred in the blade 6. At this time, the blade condition monitoring system 7 gives an acousto-optic alarm prompt to realize the online measurement of the crack of the rotating machinery blade.

Claims (1)

1. A blade tip timing and support vector machine principle-based blade crack online measurement method includes the steps that a plurality of blade tip timing sensors are arranged at different positions of a casing of a rotary machine, under a simulation working condition, the arrival time of a blade reaching each blade tip timing sensor is measured through the blade tip timing sensors, vibration displacement, amplitude and vibration frequency blade vibration data of the blade are measured, the vibration data include vibration data of a crack-free blade and a crack-containing blade, the vibration data of the crack-free blade are marked as positive samples, the vibration data of the crack-containing blade are marked as negative samples, the samples are sent to a support vector machine to be classified and trained, a classification model is built, and online measurement of cracks of the rotary machine blade under an operation working condition is achieved through the trained classification model. Comprises the following steps:

(1) arranging a plurality of blade tip timing sensors at different positions of a rotating mechanical casing, wherein the blade tip timing sensors are used for measuring the time of blades reaching each blade tip timing sensor at different rotating speeds;

(2) carrying out a simulation working condition test when the blade is intact and has no crack, sending an arrival time signal of each blade measured by the blade tip timing sensor into a blade state monitoring system, and calculating to obtain vibration displacement, amplitude, vibration frequency, initial phase, vibration constant offset and resonance frequency multiplication blade vibration data of the crack-free blade at different rotating speeds by combining the installation positions of the blade tip timing sensors, different rotating speeds of a rotating machine and a blade tip timing algorithm;

(3) for the cracked blade, carrying out a simulation working condition experiment to obtain vibration displacement, amplitude, vibration frequency, initial phase, vibration constant offset and resonance frequency multiplication frequency blade vibration data of the cracked blade at different rotating speeds;

(4) setting the vibration data of the crack-free blade as a positive sample and setting the data label as 0; setting the vibration data of the blade with the crack as a negative sample, and setting a data label as 1;

(5) taking the total number of the negative samples as reference, sampling the positive samples by using an equal interval downsampling method to realize the balance of the positive samples and the negative samples, and combining the balanced positive samples and the balanced negative samples to form a training set;

(6) normalizing the data in the training set by using a minimum and maximum normalization algorithm to eliminate dimension;

(7) according to the training set with the dimensions eliminated, carrying out classification training by using a support vector machine algorithm, and establishing a support vector machine classification model;

(8) when the rotary machine runs in a working state, according to blade arrival time obtained by actually measuring by a blade tip timing sensor, calculating by using a blade state monitoring system to obtain vibration displacement, amplitude, vibration frequency, initial phase, vibration normal deviation and resonance frequency multiplication frequency blade vibration data of the blade at different rotating speeds in the working state;

(9) the blade vibration data measured in the working state is used as a sample to be measured, the sample is sent to a trained support vector machine classification model in a blade state monitoring system after normalization processing, the label of the sample to be measured obtained through calculation of the support vector machine classification model is 0 or 1, and online measurement of the rotary machine blade cracks is achieved.

Images