Turbine blade vibration data acquisition system

Abstract

The invention relates to the technical field of blade vibration acquisition, and particularly discloses a steam turbine blade vibration data acquisition system, which comprises a region analysis module used for acquiring vibration data of blades in different regions of a certain stage of a steam turbine, determining a steep peak fusion value through parameter calculation, and determining a vibration analysis region; the abrupt change analysis module is used for processing vibration analysis area data and judging steep peak abnormal moving points and steep peak smooth difference distribution symmetry of the steep peak abnormal moving points; the area comparison module is used for comparing other areas at the same level when the vibration distribution is asymmetric, and evaluating the universality of asymmetric vibration distribution; the trend analysis module is used for drawing a steep peak fusion value change curve and judging a growth trend under a non-general condition, and analyzing the change trend of steep peak fusion values of other areas of the blade if the growth trend is presented; the method can monitor the vibration data of the steam turbine blade in multiple dimensions, and provides data support for state evaluation and fault early warning of the steam turbine blade.

Description

Technical Field

[0001] The present invention relates to the technical field of blade vibration acquisition, and particularly relates to a steam turbine blade vibration data acquisition system. Background Art

[0002] The stable operation of steam turbine blades plays an important role in the performance and economic benefits of the entire steam turbine unit. There are many deficiencies in the existing steam turbine blade vibration data acquisition technologies.

[0003] A Chinese patent application with the publication number CN108827454B discloses a method for acquiring and processing steam turbine shafting vibration data, including: synchronously acquiring multi-channel original vibration signals at high speed, preprocessing the signals and performing software resampling, extracting real-time features according to analysis requirements, standardizing the feature values and then encrypting and remotely sending them. Equal-angle and equal-time software resampling are respectively independently performed on each analysis unit of each channel; and the shaft center orbit is analyzed for the equal-angle sampled data, reducing the analysis error; when calculating the frequency spectrum, the vibration frequency spectrum of each rotation of the steam turbine is first independently calculated and then cumulatively averaged, reducing the error caused by using the same sampling rate for multiple rotations; when sending data, only the feature values required for data analysis are sent, avoiding sending the original vibration data and reducing the requirement for network bandwidth.

[0004] The prior art does not analyze the vibration data of blades in different regions of a certain stage of a steam turbine, and it is difficult to determine the regions with severe vibration and prone to failures. By obtaining the vibration data of other regions at the same stage and calculating the steep peak fusion value, the vibration analysis region can be accurately located. This helps to focus on potential fault points, improve the efficiency of fault troubleshooting, avoid blind inspections, and save time and labor costs.

[0005] When processing vibration data, the prior art lacks effective means to capture abnormal vibration mutations and cannot timely detect minor abnormal changes during the operation of steam turbine blades. If the steep peak smoothing difference is calculated through an exponential smoothing model to determine the steep peak abnormal points and judge the symmetry of the distribution of the steep peak smoothing difference, the abnormal changes in vibration can be keenly captured, such as vibration mutations caused by early cracks in the blades. This provides a key basis for early fault diagnosis.

[0006] When vibration anomalies occur, the prior art cannot determine whether the anomaly is a local problem of individual blades or a systematic problem in other regions at the same stage. When the distribution of the steep peak smoothing difference at the steep peak abnormal points is asymmetric, if the same position regions of other regions at the same stage and the vibration analysis region are used as the comparison analysis regions and the blade similarity value is quantitatively analyzed, it can accurately determine whether the asymmetry of the vibration distribution is widespread. This helps the staff quickly judge the nature of the fault, formulate a unified maintenance strategy for common problems, and conduct targeted treatment for individual problems, improving the pertinence and effectiveness of maintenance.

[0007] To this end, the present invention provides a steam turbine blade vibration data acquisition system. Summary of the Invention

[0008] The object of the present invention is to provide a steam turbine blade vibration data acquisition system to solve the above background problems.

[0009] The object of the present invention can be achieved by the following technical solutions:

[0010] A steam turbine blade vibration data acquisition system includes the following modules:

[0011] Region analysis module: used to obtain the vibration data of blades in different regions of a certain stage of the steam turbine, perform numerical analysis, obtain the steep peak fusion value, and determine the vibration analysis region of the steam turbine blade;

[0012] Mutation analysis module: used to perform numerical analysis on the steep peak fusion values of multiple monitoring periods within the vibration analysis region, determine the steep peak abnormal points within the monitoring period, calculate the steep peak smoothing difference of the steep peak abnormal points, and judge whether the distribution of the steep peak smoothing differences of the steep peak abnormal points is symmetric;

[0013] Region comparison module: if the distribution is asymmetric, used to take other regions of the same stage of the steam turbine as the comparison analysis region, perform quantitative analysis on the vibrations of the comparison analysis region and the vibration analysis region, and judge whether the asymmetric vibration distribution in the vibration analysis region is a common phenomenon;

[0014] Trend analysis module: if it is not a common phenomenon, perform trend analysis on the steep peak fusion value of the vibration analysis region, judge whether the steep peak fusion value shows an increasing trend in the time dimension, and if it shows an increasing trend, judge whether the increasing trend of the steep peak fusion value is limited to the vibration analysis region of the steam turbine blade.

[0015] As a further technical solution of the present invention: the determination method of the vibration analysis region of the steam turbine blade is as follows:

[0016] Obtain the initial vibration data of the steam turbine blade, divide the collected initial vibration data into multiple monitoring periods, and obtain the amplitude at each monitoring moment within the monitoring period;

[0017] Based on the amplitudes at all monitoring moments within the monitoring period, perform numerical analysis on the amplitude at each monitoring moment to obtain the steep peak fusion value;

[0018] Obtain the steep peak fusion values of different regions of the steam turbine blade, sort them, and determine the region of the steam turbine blade corresponding to the maximum steep peak fusion value as the vibration analysis region.

[0019] As a further technical solution of the present invention: the method for obtaining the steep peak fusion value is as follows:

[0020] Calculate the kurtosis of the monitoring period through the kurtosis formula, obtain the root mean square value of the monitoring period, obtain the maximum value of the amplitude within the monitoring period, and obtain the period peak value;

[0021] Perform a ratio process on the period peak value and the root mean square value within the monitoring period to obtain the peak factor;

[0022] Perform a weighted summation process on the peak factor and the kurtosis of the amplitude to obtain the steep peak fusion value.

[0023] As a further technical solution of the present invention: the acquisition method of the steep peak abnormal point is:

[0024] Based on the vibration analysis area, obtain the steep peak fusion values within multiple monitoring periods;

[0025] Take the steep peak fusion value of the first monitoring period as the initial smoothing value, and based on the initial smoothing value, obtain the exponential smoothing value of each monitoring period through the exponential smoothing formula;

[0026] Perform a difference process on the steep peak fusion value and the exponential smoothing value of each monitoring period to obtain the steep peak smoothing difference;

[0027] If the steep peak smoothing difference is higher than the preset smoothing difference threshold, mark the time point corresponding to the monitoring period as the steep peak abnormal point.

[0028] As a further technical solution of the present invention: the judgment method for whether the distribution of the steep peak smoothing difference of the steep peak abnormal point is symmetric is:

[0029] By calculating the skewness of the steep peak smoothing difference corresponding to the steep peak abnormal point, judge whether the distribution of the steep peak smoothing differences of the steep peak abnormal points within multiple monitoring periods is symmetric;

[0030] Obtain the skewness of the steep peak smoothing difference of the steep peak abnormal point through the skewness formula;

[0031] If the skewness of the steep peak smoothing difference is 0, it is considered that the distribution of the steep peak smoothing difference is symmetric; otherwise, the distribution of the steep peak smoothing differences of the steep peak abnormal points is asymmetric.

[0032] As a further technical solution of the present invention: the judgment method for whether the vibration distribution asymmetry in the vibration analysis area is a common phenomenon is:

[0033] If the distribution of the steep peak smoothing differences of the steep peak abnormal points is asymmetric, use other areas of the same stage of the steam turbine as the comparison analysis area;

[0034] Obtain the steep peak fusion values in the comparison analysis area within the monitoring period and construct a comparison analysis group;

[0035] Obtain the monitoring period and the steep peak fusion values of the current steam turbine blade vibration analysis area and construct a vibration analysis group;

[0036] Based on the sharp peak fusion value of the vibration analysis group and the sharp peak fusion value of the comparison analysis group, similarity analysis is performed to obtain the vibration similarity region;

[0037] Quantitative analysis is performed on the vibration similarity region to obtain the blade similarity value;

[0038] If the blade similarity value is higher than the preset blade similarity threshold, the asymmetric vibration distribution in the vibration analysis region is a common phenomenon.

[0039] As a further technical solution of the present invention: The obtaining method of the vibration similarity region is:

[0040] Through the mutual information formula, obtain the similarity information value between the vibration analysis group and the comparison analysis group;

[0041] If the similarity information value between the vibration analysis group and the comparison analysis group is higher than the preset similarity information threshold, it is considered that the vibration analysis group and the comparison analysis group are similar in the time dimension distribution;

[0042] If the comparison analysis region and the vibration analysis region are similar in the time dimension distribution, mark the comparison analysis region as the vibration similarity region.

[0043] As a further technical solution of the present invention: The obtaining method of the blade similarity value is:

[0044] Obtain the number of vibration similarity regions in all comparison analysis regions to obtain the number of similarity regions;

[0045] Obtain the number of all comparison analysis regions, and perform a ratio process on the number of similarity regions and the number of all comparison analysis regions to obtain the similarity region ratio;

[0046] Obtain the similarity information value of the vibration similarity region, and perform a difference process on the similarity information value of the vibration similarity region and the preset similarity information threshold to obtain the similarity information difference;

[0047] Sum and average the similarity information differences of all vibration similarity regions to obtain the similarity information average value;

[0048] Perform a ratio process on the similarity information average value and the preset similarity information threshold to obtain the similarity proximity ratio;

[0049] Based on the similarity region ratio and the similarity proximity ratio, calculate the blade similarity value through the weighted formula.

[0050] As a further technical solution of the present invention: The determination method of whether the growth trend of the sharp peak fusion value is limited to the vibration analysis region of the steam turbine blade is:

[0051] Obtain the vibration analysis group of the vibration analysis area, extract the steep peak fusion value of the vibration analysis group, use the monitoring period as the X-axis and the steep peak fusion value as the Y-axis, and plot the steep peak change curve in the two-dimensional rectangular coordinate system;

[0052] Conduct trend analysis on the steep peak change curve. If the steep peak change curve shows an increasing trend in the time dimension, obtain the steep peak fusion values of other areas at the same level on the steam turbine blade where the vibration analysis area is located;

[0053] Calculate the steep peak fusion values of other areas at the same level. If the steep peak fusion values of other areas also show an increasing trend in the time dimension, it is considered that the increasing trend of the steep peak fusion value is not limited to the initial vibration analysis area.

[0054] As a further technical solution of the present invention: The determination method for the steep peak change curve showing an increasing trend in the time dimension is as follows:

[0055] Obtain each inflection point on the steep peak change curve, use the inflection point as the demarcation point of the curve, and divide the steep peak change curve into multiple curve segments;

[0056] Obtain the difference between the steep peak fusion values of the two endpoints corresponding to the curve segment to get the endpoint steep peak difference. If the endpoint steep peak difference is positive, mark the curve segment as an increasing curve segment, otherwise mark it as a decreasing curve segment;

[0057] Obtain the number of increasing curve segments and the number of all curve segments, and perform a ratio process on the number of increasing curve segments and the number of all curve segments to obtain the increasing segment ratio;

[0058] Obtain the number of decreasing curve segments and the number of all curve segments, and perform a ratio process on the number of decreasing curve segments and the number of all curve segments to obtain the decreasing segment ratio;

[0059] Perform a difference process on the increasing segment ratio and the decreasing segment ratio, and perform a weighted summation process on the obtained result and the endpoint steep peak difference to obtain the growth determination value;

[0060] Compare the growth determination value with the preset growth determination threshold. If the growth determination value is higher than the preset growth determination threshold, the steep peak fusion value and the corresponding steep peak change curve show an increasing trend in the time dimension.

[0061] The beneficial effects of the present invention:

[0062] (1) By acquiring the vibration data of blades in different regions at a certain level, calculating the steep peak fusion value and determining the vibration analysis region, comprehensively analyzing the vibration amplitude fluctuation and distribution steepness, it is possible to focus on the regions with the most intense vibration and the most prone to problems. Compared with traditional acquisition methods, potential faults can be quickly detected. By calculating the steep peak smoothing difference using the exponential smoothing model, determining the steep peak abnormal points, and judging the symmetry of the steep peak smoothing difference distribution, random noise and short-term fluctuations can be removed, and abnormal changes during the operation of steam turbine blades can be captured in a timely manner, such as the vibration mutation when a blade has a crack, providing key information for early fault diagnosis, facilitating staff to take measures in a timely manner to avoid the expansion of faults.

[0063] (2) When the distribution of the steep peak smoothing difference at the steep peak abnormal points is asymmetric, other regions at the same level are used as the comparison and analysis regions, and the vibration data is quantitatively analyzed to obtain the blade similarity value to judge whether the asymmetric vibration distribution is common. This helps to distinguish the local problems of individual blades from the systematic problems of the entire steam turbine blades, enabling staff to more accurately evaluate the nature of the faults and adopt targeted maintenance strategies to reduce maintenance costs. When the asymmetric vibration distribution is not a common phenomenon, a trend analysis is performed on the steep peak fusion value of the vibration analysis region to judge whether its growth trend is limited to that region. Analyze the development trend of the steam turbine blade vibration from multiple dimensions, discover the systematic problems existing in the steam turbine blades in advance, provide a scientific basis for the maintenance and management of the steam turbine, reasonably arrange the maintenance plan, and ensure the stable operation of the steam turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The present invention will be further described below with reference to the accompanying drawings.

[0065] Figure 1 is a module diagram of a steam turbine blade vibration data acquisition system of the present invention;

[0066] Figure 2 is a flowchart of a method for obtaining similarity information values provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0068] Embodiment 1

[0069] Please refer to Figure 1 As shown, the present invention is a steam turbine blade vibration data acquisition system, including the following modules:

[0070] Region analysis module: It is used to obtain the vibration data of blades in different regions of a certain stage of a steam turbine, extract the characteristic values of the vibration data and perform numerical analysis to obtain the steep peak fusion value, and determine the vibration analysis region of the steam turbine blades;

[0071] In some embodiments, according to the different functions and positions of the steam turbine blades, the blades of the steam turbine are divided into multiple levels, including: governing stage blades, pressure stage blades, and last stage blades. Vibration sensors are installed at different levels and different regions of the steam turbine to collect the vibration data of the blades in different regions of a certain stage of the steam turbine, and the initial vibration data is obtained;

[0072] Divide the collected initial vibration data into multiple monitoring periods, obtain the amplitude at each monitoring moment within the monitoring period, and mark the amplitude as x i , where i is the number of the monitoring moment, and the value range of i is [1, n];

[0073] Based on the amplitudes at all monitoring moments within the monitoring period, calculate the kurtosis qd of the monitoring period amplitude;

[0074] Specifically, through the formula: Obtain the kurtosis qd of the amplitude of each monitoring period;

[0075] Where represents the fourth-order central moment of the amplitude within the monitoring period, represents the second-order central moment of the amplitude within the monitoring period, that is, the variance;

[0076] It should be noted that kurtosis is used to describe the steepness of the vibration amplitude distribution. When there is an impact or abnormality in the vibration of the steam turbine blades, the kurtosis value will increase significantly, which is used to detect whether there are sudden faults or abnormal vibrations. When cracks appear in the blades, etc., the kurtosis value may change significantly;

[0077] Through the formula: Obtain the root mean square value RMS of the monitoring period, obtain the maximum value of the amplitude within the monitoring period, and obtain the period peak value;

[0078] Perform a ratio process on the period peak value and the root mean square value within the monitoring period to obtain the peak factor;

[0079] Perform a weighted summation process on the peak factor and the kurtosis of the amplitude to obtain the steep peak fusion value;

[0080] Obtain the steep peak fusion values of different regions of the steam turbine blades, arrange them in descending order, and determine the steam turbine blade region corresponding to the maximum steep peak fusion value as the vibration analysis region;

[0081] It should be noted that the steep peak fusion value comprehensively reflects the amplitude fluctuation and distribution steepness of the vibration of the steam turbine blade, and more comprehensively reflects the characteristics of the vibration of the steam turbine blade. By comparing the steep peak fusion values in different regions, the region corresponding to the maximum value is determined as the vibration analysis region, which can focus on the region with the most intense vibration and the most likely problems;

[0082] Mutation analysis module: used to perform numerical analysis on the steep peak fusion values of multiple monitoring periods within the vibration analysis region, obtain the steep peak smoothing difference, determine the steep peak abnormal points within the monitoring period, and analyze the distribution of the steep peak smoothing differences of the steep peak abnormal points to determine whether the distribution of the steep peak smoothing differences of the steep peak abnormal points is symmetric;

[0083] Based on the vibration analysis region, obtain the steep peak fusion values within multiple monitoring periods, and mark the steep peak fusion value of each monitoring period as qd t , where t is the number of the monitoring period;

[0084] Based on the exponential smoothing model, determine the steep peak abnormal points of the vibration analysis region in all monitoring periods;

[0085] Specifically, take the steep peak fusion value of the first monitoring period as the initial smoothing value S1;

[0086] Through the formula: S t =α*qd t +(1 - α)*S t-1 Obtain the exponential smoothing value S of each monitoring period t , where α is the smoothing coefficient, and the value range of α is (0, 1);

[0087] Preferably, α is 0.65;

[0088] It should be noted that the exponential smoothing model calculates the exponential smoothing value, which can effectively remove the random noise and short-term fluctuations in the original steep peak fusion value. By calculating the steep peak smoothing difference and comparing it with the preset threshold, it can detect whether there are steep peak abnormal points within the monitoring period, that is, the moments or regions where the vibration state changes suddenly, which helps to capture the abnormal changes in the operation of the steam turbine blade in a timely manner;

[0089] Take the difference between the steep peak fusion value qd of each monitoring period t and the exponential smoothing value S t to obtain the steep peak smoothing difference;

[0090] Compare the steep peak smoothing difference with the preset smoothing difference threshold to determine the steep peak abnormal points of the monitoring period;

[0091] If the steep peak smoothing difference is higher than the preset smoothing difference threshold, mark the time point corresponding to the monitoring period as the steep peak abnormal point;

[0092] By calculating the steep peak anomaly point and the corresponding steep peak smoothing error skewness, it is determined whether the steep peak smoothing error distribution of the steep peak anomaly point in multiple monitoring periods is symmetrical;

[0093] By formula: Get the steep peak smoothing difference skewness SK of the steep peak anomaly point, where e t , Se t , m represent the steep peak smoothing difference of the steep peak anomaly points in each monitoring period, the mean of the steep peak smoothing difference of the steep peak anomaly points in m monitoring periods, and the standard deviation of the steep peak smoothing difference of the steep peak anomaly points in m monitoring periods, respectively, and m is the total number of monitoring periods;

[0094] If SK = 0, it means that the distribution of the steep peak smoothing difference in multiple monitoring periods is symmetrical, that is, the distribution of the steep peak smoothing difference on both sides of the mean is relatively uniform, and there is no obvious bias;

[0095] If SK>0 or SK<0, the distribution of the steep peak smoothing difference of the steep peak anomaly point is asymmetric. SK>0 means that the distribution of the steep peak smoothing difference in multiple monitoring periods is positively skewed, which means that the probability of a large positive steep peak smoothing difference is relatively small.

[0096] SK<0, indicating that the distribution of steep peak smoothing differences in multiple monitoring periods is negatively skewed, which means that the probability of a large negative steep peak smoothing difference is relatively small;

[0097] The technical solution of this embodiment is: obtaining vibration data of blades in different areas of a certain stage of a steam turbine, extracting characteristic values ​​of the vibration data and performing numerical analysis to obtain steep peak fusion values, determining the vibration analysis area of ​​the steam turbine blades, performing numerical analysis on the steep peak fusion values ​​of multiple monitoring periods within the vibration analysis area to obtain steep peak smoothing differences, determining steep peak anomaly points within the monitoring period, and analyzing the steep peak smoothing difference distribution of the steep peak anomaly points to determine whether the steep peak smoothing difference distribution of the steep peak anomaly points is symmetrical, thereby providing more dimensional data support for further evaluating vibration anomalies and facilitating timely discovery of potential fault hazards.

[0098] Example 2

[0099] like Figure 1 As shown, a steam turbine blade vibration data acquisition system also includes the following modules

[0100] Regional comparison module: If the steep peak smoothing difference distribution of the steep peak anomaly point is asymmetric, it is used to obtain other regions of the same level as the comparison analysis region, and the vibration of the comparison analysis region and the vibration analysis region is quantitatively analyzed to obtain the blade similarity value, and to determine whether the asymmetric vibration distribution in the vibration analysis region is a common phenomenon;

[0101] If the distribution of the steep peak smoothing difference of the steep peak abnormal points is asymmetric, other areas at the same level are used as the analysis and comparison areas;

[0102] Obtain the steep peak fusion value of the comparison and analysis area during the monitoring period, and construct a comparison and analysis group;

[0103] Obtain the steep peak fusion value of the current steam turbine blade vibration analysis area during the monitoring period, and construct a vibration analysis group;

[0104] As Figure 2 shown, based on the steep peak fusion value of the vibration analysis group and the steep peak fusion value of the comparison and analysis group, the similarity information value Ph between the vibration analysis group and the comparison and analysis group is obtained through the mutual information formula;

[0105] It should be noted that since there are multiple comparison and analysis areas, there are also multiple comparison and analysis groups;

[0106] Specifically, obtain the value range of the steep peak fusion value in the vibration analysis group and the value range of the steep peak fusion value in the comparison and analysis group;

[0107] Divide the value range of the steep peak fusion value of the vibration analysis group into multiple equal-width intervals to obtain vibration sub-intervals, and the number of each vibration sub-interval is k1;

[0108] Divide the value range of the steep peak fusion value of the comparison and analysis group into multiple equal-width intervals to obtain comparison sub-intervals, and the number of each comparison sub-interval is k2;

[0109] Exemplarily, if the number of the vibration sub-interval is 31, it means the third vibration sub-interval; if the number of the comparison sub-interval is 12, it means the first comparison sub-interval;

[0110] Calculate the probability that the steep peak fusion value in the vibration analysis group falls into each vibration sub-interval to obtain P(k1);

[0111] Calculate the probability that the steep peak fusion value in the comparison and analysis group falls into each comparison sub-interval to obtain P(k2);

[0112] Calculate the joint probability P(k1,k2) that the steep peak fusion values in the comparison and analysis group and the vibration analysis group fall into the vibration sub-interval and the comparison sub-interval at the same time;

[0113] Exemplarily, if there are 200 steep peak fusion value data in the vibration analysis group and the comparison and analysis group respectively, divide the vibration analysis group into 4 vibration sub-intervals, which are [0,20], (20,40], (40,60], (60,80];

[0114] The comparison analysis group is divided into 4 comparison sub-intervals, namely (10, 30], (30, 50], (50, 70], and (70, 90]. At the same time, if the number of steep peak fusion values where the vibration analysis group falls within (20, 40] and the comparison analysis group falls within (30, 50] is 20, then the joint probability P(21, 22) = 20 / 200 = 0.1;

[0115] Through the mutual information formula: Obtain the similarity information value Ph between the vibration analysis group and the comparison analysis group;

[0116] If the similarity information value Ph between the vibration analysis group and the comparison analysis group is higher than the preset similarity information threshold, it is considered that the vibration analysis group and the comparison analysis group are similar in distribution in the time dimension. Otherwise, they are not similar;

[0117] If the comparison analysis area is similar to the vibration analysis area, mark the comparison analysis area as a vibration similar area;

[0118] Obtain the number of vibration similar areas in all comparison analysis areas to get the number of similar areas;

[0119] Obtain the number of all comparison analysis areas, and perform a ratio process on the number of similar areas and the number of all comparison analysis areas to get the similarity area ratio;

[0120] Obtain the similarity information value of the vibration similar area, and perform a difference process on the similarity information value of the vibration similar area and the preset similarity information threshold to get the similarity information difference;

[0121] Sum up and average the similarity information differences of all vibration similar areas to get the similarity information average value;

[0122] Perform a ratio process on the similarity information average value and the preset similarity information threshold to get the similarity proximity ratio;

[0123] Obtain the blade similarity value Xz through the formula: Xz = a1*Xq + a2*Xs, where a1 and a2 take 0.36 and 0.84 respectively, and Xq and Xs represent the similarity area ratio and the similarity proximity ratio respectively;

[0124] Compare the blade similarity value with the preset blade similarity threshold to determine whether the vibration distribution asymmetry in the vibration analysis area widely exists in other areas at the same level;

[0125] If the blade similarity value is higher than the preset blade similarity threshold, it is considered that the vibration distribution asymmetry in the vibration analysis area generally exists in other areas at the same level, that is, the vibration distribution asymmetry in the vibration analysis area is a common phenomenon;

[0126] If the blade similarity value is higher than the preset blade similarity threshold, it is considered that the asymmetric vibration distribution in the vibration analysis area is an individual phenomenon in other areas of the same level;

[0127] Trend analysis module: If the asymmetric vibration distribution in the vibration analysis area is not a common phenomenon, it is used to perform trend analysis on the steep peak fusion value in the vibration analysis area to determine whether the steep peak fusion value shows an increasing trend in the time dimension. If it shows an increasing trend, it is determined whether the increasing trend of the steep peak fusion value is limited to the vibration analysis area of the steam turbine blade;

[0128] Obtain the vibration analysis group in the vibration analysis area, extract the steep peak fusion value of the vibration analysis group, use the monitoring period as the X-axis and the steep peak fusion value as the Y-axis, and plot the steep peak change curve in the two-dimensional rectangular coordinate system;

[0129] Obtain each inflection point on the steep peak change curve, use the inflection point as the demarcation point of the curve, and divide the steep peak change curve into multiple curve segments;

[0130] Obtain the difference between the steep peak fusion values of the two endpoints corresponding to the curve segment to get the endpoint steep peak difference. If the endpoint steep peak difference is positive, mark the curve segment as an increasing curve segment, otherwise mark it as a decreasing curve segment;

[0131] Obtain the number of increasing curve segments and the number of all curve segments, and perform a ratio process on the number of increasing curve segments and the number of all curve segments to get the increasing segment ratio;

[0132] Obtain the number of decreasing curve segments and the number of all curve segments, and perform a ratio process on the number of decreasing curve segments and the number of all curve segments to get the decreasing segment ratio;

[0133] Perform a difference process on the increasing segment ratio and the decreasing segment ratio, and perform a weighted summation process on the obtained result and the endpoint steep peak difference to get the growth determination value;

[0134] Compare the growth determination value with the preset growth determination threshold. If the growth determination value is higher than the preset growth determination threshold, the steep peak fusion value and the corresponding steep peak change curve show an increasing trend in the time dimension. Otherwise, the steep peak fusion value and the corresponding steep peak change curve show a decreasing trend in the time dimension;

[0135] If the steep peak change curve shows an increasing trend, obtain the steep peak fusion values of other areas on the steam turbine blade where the vibration analysis area is located;

[0136] Calculate the steep peak fusion values of other areas at the same level. If the steep peak fusion values of other areas also show an increasing trend in the time dimension, it is considered that the increasing trend of the steep peak fusion value is not limited to the initial vibration analysis area;

[0137] The technical solution of this embodiment is as follows: If the steep peak smoothing difference distribution of the steep peak abnormal points is asymmetric, other areas at the same level are used as the comparison and analysis areas, and the vibrations of the comparison and analysis areas and the vibration analysis areas are quantitatively analyzed to obtain the blade similarity value, and it is judged whether the asymmetric vibration distribution in the vibration analysis area is a common phenomenon. If it is not a common phenomenon, the trend analysis of the steep peak fusion value in the vibration analysis area is used to judge whether the steep peak fusion value shows an increasing trend in the time dimension. If it shows an increasing trend, it is judged whether the increasing trend of the steep peak fusion value is limited to the vibration analysis area of the steam turbine blade, which is beneficial to analyzing the development trend of the steam turbine blade vibration, timely discovering possible systematic problems, and providing a basis for the maintenance and management of the steam turbine.

[0138] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the present invention.

Claims

1. A steam turbine blade vibration data acquisition system, characterized in that: It includes the following modules: Region analysis module: It is used to obtain the vibration data of blades in different regions of a certain stage of the steam turbine, perform numerical analysis, obtain the steep peak fusion value, and determine the vibration analysis region of the steam turbine blades; Mutation analysis module: It is used to perform numerical analysis on the steep peak fusion values of multiple monitoring periods within the vibration analysis region, determine the steep peak abnormal points within the monitoring period, calculate the steep peak smoothing difference of the steep peak abnormal points, and judge whether the distribution of the steep peak smoothing difference of the steep peak abnormal points is symmetric; Region comparison module: If the distribution is asymmetric, it is used to take other regions of the same stage as the comparison analysis region, perform quantitative analysis on the vibration data of the comparison analysis region and the vibration analysis region, and judge whether the asymmetric vibration distribution of the vibration analysis region is a common phenomenon; Trend analysis module: If it is not a common phenomenon, perform trend analysis on the steep peak fusion value of the vibration analysis region, judge whether the steep peak fusion value shows an increasing trend in the time dimension, and if it shows an increasing trend, judge whether the increasing trend of the steep peak fusion value is limited to the vibration analysis region of the steam turbine blades.

2. The steam turbine blade vibration data acquisition system according to claim 1, characterized in that: The determination method of the vibration analysis region of the steam turbine blades is as follows: Obtain the initial vibration data of the steam turbine blades, divide the collected initial vibration data into multiple monitoring periods, and obtain the amplitude at each monitoring moment within the monitoring period; Based on the amplitudes at all monitoring moments within the monitoring period, perform numerical analysis on the amplitude at each monitoring moment to obtain the steep peak fusion value; Obtain the steep peak fusion values of different regions of the steam turbine blades, sort them, and determine the region of the steam turbine blade corresponding to the maximum steep peak fusion value as the vibration analysis region.

3. The steam turbine blade vibration data acquisition system according to claim 2, characterized in that: The acquisition method of the steep peak fusion value is as follows: Calculate the kurtosis of the amplitude within the monitoring period through the kurtosis formula, obtain the root mean square value of the monitoring period and the maximum value of the amplitude within the monitoring period to obtain the period peak value; Perform ratio processing on the period peak value within the monitoring period and the root mean square value to obtain the peak factor; Perform weighted summation processing on the peak factor and the kurtosis of the amplitude to obtain the steep peak fusion value.

4. A steam turbine blade vibration data acquisition system according to claim 1, characterized in that: The acquisition method of the steep peak abnormal point is as follows: Based on the vibration analysis region, obtain the steep peak fusion values within multiple monitoring periods; Take the steep peak fusion value of the first monitoring period as the initial smoothing value, and based on the initial smoothing value, obtain the exponential smoothing value of each monitoring period through the exponential smoothing formula; Perform difference processing on the steep peak fusion value of each monitoring period and the exponential smoothing value to obtain the steep peak smoothing difference; If the steep peak smoothing difference is higher than the preset smoothing difference threshold, mark the time point corresponding to the monitoring period as the steep peak abnormal point.

5. A steam turbine blade vibration data acquisition system according to claim 1, characterized in that: The judgment method for whether the distribution of the steep peak smoothing difference of the steep peak abnormal point is symmetric is as follows: Obtain the skewness of the steep peak smoothing difference of the steep peak abnormal point through the skewness formula; If the skewness of the steep peak smoothing difference is 0, it is considered that the distribution of the skewness of the steep peak smoothing difference is symmetric, otherwise, the distribution of the steep peak smoothing difference of the steep peak abnormal point is asymmetric.

6. The steam turbine blade vibration data acquisition system according to claim 1, wherein: The judgment method for whether the asymmetric vibration distribution of the vibration analysis region is a common phenomenon is as follows: If the distribution of the steep peak smoothing difference of the steep peak abnormal point is asymmetric, take other regions of the same stage as the comparison analysis region; Obtain the steep peak fusion values of the comparison analysis region within the monitoring period and construct a comparison analysis group; Obtain the monitoring period, the steep peak fusion value of the current steam turbine blade vibration analysis area, and construct a vibration analysis group; Based on the steep peak fusion value of the vibration analysis group and the steep peak fusion value of the comparison analysis group, conduct a similarity analysis to obtain a vibration similarity area; Conduct a quantitative analysis on the vibration similarity area to obtain a blade similarity value; If the blade similarity value is higher than the preset blade similarity threshold, the vibration distribution in the vibration analysis area is asymmetric and is a common phenomenon.

7. The steam turbine blade vibration data acquisition system according to claim 6, wherein: The method for obtaining the vibration similarity area is as follows: Through the mutual information formula, obtain the similarity information value between the vibration analysis group and the comparison analysis group; If the similarity information value between the vibration analysis group and the comparison analysis group is higher than the preset similarity information threshold, it is considered that the vibration analysis group and the comparison analysis group are similar in distribution in the time dimension; If the comparison analysis area and the vibration analysis area are similar in distribution in the time dimension, mark the comparison analysis area as the vibration similarity area.

8. A steam turbine blade vibration data acquisition system according to claim 7, characterized in that: The method for obtaining the blade similarity value is as follows: Obtain the number of vibration similarity areas in all comparison analysis areas to obtain the number of similarity areas; obtain the number of all comparison analysis areas, and perform a ratio process on the number of similarity areas and the number of all comparison analysis areas to obtain a similarity area ratio; Obtain the similarity information value of the vibration similarity area, and perform a difference process on the similarity information value of the vibration similarity area and the preset similarity information threshold to obtain a similarity information difference; Perform a summation and averaging process on the similarity information differences of all vibration similarity areas to obtain a similarity information average value; Perform a ratio process on the similarity information average value and the preset similarity information threshold to obtain a similarity proximity ratio; Based on the similarity area ratio and the similarity proximity ratio, calculate the blade similarity value through a weighted formula.

9. The steam turbine blade vibration data acquisition system according to claim 6, characterized in that: The determination method for whether the growth trend of the steep peak fusion value is limited to the vibration analysis area of the steam turbine blade is as follows: Obtain the vibration analysis group of the vibration analysis area, extract the steep peak fusion value of the vibration analysis group, use the monitoring period as the X-axis and the steep peak fusion value as the Y-axis, and draw a steep peak change curve in a two-dimensional rectangular coordinate system; Conduct a trend analysis on the steep peak change curve. If the steep peak change curve shows an increasing trend in the time dimension, obtain the steep peak fusion values of other areas of the steam turbine blade where the vibration analysis area is located; Calculate the steep peak fusion values of other areas in the same flow path of the steam turbine blade. If the steep peak fusion values of other areas also show an increasing trend in the time dimension, it is considered that the growth trend of the steep peak fusion value is not limited to the initial vibration analysis area.

10. A steam turbine blade vibration data acquisition system according to claim 9, characterized in that: The determination method for the steep peak change curve showing an increasing trend in the time dimension is as follows: Obtain each inflection point on the steep peak change curve, use the inflection point as the boundary point of the curve, and divide the steep peak change curve into multiple curve segments; Obtain the difference between the steep peak fusion values of the two endpoints corresponding to the curve segment to obtain an endpoint steep peak difference. If the endpoint steep peak difference is positive, mark the curve segment as an increasing curve segment, otherwise mark it as a decreasing curve segment; Obtain the number of increasing curve segments and the number of all curve segments, and perform a ratio process on the number of increasing curve segments and the number of all curve segments to obtain an increasing segment ratio; Obtain the number of decreasing curve segments and the number of all curve segments, and perform a ratio process on the number of decreasing curve segments and the number of all curve segments to obtain a decreasing segment ratio; Perform a difference operation on the growth line segment ratio and the decline line segment ratio, and perform a weighted summation operation on the obtained result and the endpoint peak difference to obtain a growth determination value; Compare the growth determination value with a preset growth determination threshold. If the growth determination value is higher than the preset growth determination threshold, the peak fusion value and the corresponding peak change curve show a growth trend in the time dimension.

Images