A method for detecting vibration tolerance of spacer rods

Abstract

The present invention relates to the field of data processing, and more specifically, to a method for detecting the vibration tolerance of a spacer bar. The method comprises: obtaining the amplitude value of a conductor and the stress value at a connection point of the spacer bar corresponding to the conductor based on a time series, and obtaining the change credibility between the amplitude value and the stress value at each moment in history; fitting a curve to the data points of the stress value at all moments according to the least square method, so as to calculate the deviation degree of the stress value at all moments; adopting an isolation forest algorithm, combining the change credibility and the deviation degree, obtaining the priority of the segmentation point, determining the best segmentation point, efficiently constructing an isolation tree and accurately identifying abnormal stress data, calculating the tolerance of the spacer bar vibration according to the abnormal stress data, and comparing with a preset threshold value to evaluate the state of the spacer bar. The present invention ensures the safe and stable operation of the power system by improving the segmentation efficiency and accuracy of the model.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to a method for detecting the vibration tolerance of a spacer rod. Background Art

[0002] Spacers are an important hardware used in high-voltage transmission lines. They are mainly used to maintain a fixed distance between conductors to prevent interphase short circuits caused by conductor swaying due to natural conditions such as wind vibration and ice flash, while reducing the occurrence of air arcs to ensure the stable operation of transmission lines. Especially in harsh climatic conditions and geographical environments, the vibration load on the spacers increases significantly. In order to ensure that the spacers can continue to play their role in long-term operation, accurate detection of their vibration tolerance becomes particularly important.

[0003] The existing Chinese patent application document with publication number CN111208017A discloses a fatigue tolerance loading test device for spacer bars; it includes a conductor pre-tightening fixture and a conductor loading fixture; the conductor pre-tightening fixture and the conductor loading fixture respectively tension the left and right ends of the horizontal conductor, and are fixedly installed symmetrically along the left and right axes of the horizontal direction of the conductor; the tensioned conductor between the conductor pre-tightening fixture and the conductor loading fixture is symmetrically installed with the six-split spacer bars to be tested; the middle part of the conductor in the middle of the six-split spacer bars is fixedly installed with a six-split loading fixture; the left and right sides of the six-split loading fixture are respectively connected to a servo vertical loading mechanism and a servo torsion loading mechanism; the servo vertical loading mechanism is used to provide a vertical vibration torque to the spacer bars on the conductor; the servo torsion loading mechanism is used to provide a torsion torque to the spacer bars on the conductor. This application document can simulate the torsional dancing of natural wind and the vibration tolerance test of the spacer plate, thereby ensuring the safety of the transmission line and the stability of the power system.

[0004] The above patent application documents require that the spacer rods themselves have sufficient mechanical strength, and that the fatigue tolerance characteristics of the spacer rods in a simulated environment be quantitatively tested and analyzed. Currently, the tolerance capacity of the spacer rods can be obtained by obtaining the anomalies of the amplitude and stress values ​​of the spacer rods. The isolation forest model is usually used to analyze the data. However, in actual scenarios, since the transmission lines are affected by different wind forces, the data will have multiple trend changes. The traditional isolation forest algorithm often randomly selects the segmentation threshold, and cannot effectively divide the inherent laws of the distribution of these different trend data, resulting in difficulty in identifying real anomalies, thereby reducing the accuracy of anomaly detection. Summary of the invention

[0005] In order to solve the problem that when traditional isolation forest obtains abnormal data values, the segmentation threshold is usually randomly selected, which cannot effectively divide the inherent laws of the distribution of these different trend data, resulting in difficulty in identifying real anomalies, thereby reducing the accuracy of anomaly detection, the present invention provides solutions in the following aspects.

[0006] A method for detecting the vibration tolerance of a spacer bar comprises: obtaining the amplitude value of a conductor and the stress value of a connection point of the conductor corresponding to the spacer bar based on a time sequence, obtaining the change credibility between the amplitude value and the stress value at each moment in history; fitting a curve to the data points of the stress value at all moments according to the least square method, so as to calculate the deviation degree of the stress value at all moments; using an isolation forest algorithm, obtaining the priority of a segmentation point according to the change credibility and deviation degree at each moment, obtaining the best segmentation point of the current node by using the priority, and recursively completing the construction of an isolation tree; obtaining abnormal stress data according to the constructed isolation forest, calculating the vibration tolerance of the spacer bar according to the proportion of abnormal stress data points and the sum of the deviations between the stress values ​​of all abnormal stress data points and the adjustable maximum stress value of the spacer bar, and judging the use status of the spacer bar according to the tolerance and a preset tolerance threshold; wherein the priority satisfies the following relationship: , where Indicates The priority of the time division point, Indicates Credibility changes all the time. Indicates The degree of deviation of the stress value at the moment, Indicates An exponential function with base .

[0007] The effect is that by comprehensively utilizing time series data collection, least squares fitting and isolation forest algorithm, an efficient and accurate evaluation of the vibration tolerance of the spacer bars in the power system is achieved. This method can not only monitor and analyze the changes in amplitude and stress values ​​in real time, but also effectively identify potential abnormal data points by calculating the credibility of the changes and the degree of deviation of the stress values ​​at each moment. It innovatively applies the priority formula to select the best split point, further optimizing the construction process of the isolation forest model. By comparing with the preset tolerance threshold, this method can timely judge the use status of the spacer bars, thereby enhancing the reliability and safety of the power system, reducing unexpected failures, reducing maintenance costs, and ensuring the continuity and stability of power supply.

[0008] Preferably, the change credibility includes:

[0009] The slope difference between the amplitude value and stress value data at each moment in history and the previous moment is calculated to obtain the change trend of the historical amplitude value and the change trend of the corresponding stress value;

[0010] The changing trends of the historical amplitude values ​​and the corresponding stress values ​​are standardized respectively, and the standardized differences between the two at all historical moments are summed up. The output range of the summed data is adjusted using the inverse tangent function, and normalized to obtain the credibility of the change at the current moment.

[0011] The effect is: by calculating the slope difference between the amplitude and stress value data at each moment and the previous moment, the changing trends of the amplitude and stress values ​​are obtained respectively, and the changing trends are standardized to ensure that data of different scales and dimensions can be compared under the same standard, thereby enhancing the applicability and flexibility of the method. The standardization and normalization processing are more sensitive to outliers, which is conducive to detecting abnormal situations.

[0012] Preferably, the change credibility further includes:

[0013] Take the current time as the target time and get the time before the target time The amplitude and stress values ​​within the target time are calculated by the exponential moving average method. The amplitude and stress values ​​within , set the smoothing factors of the amplitude value and the stress value respectively, obtain the change trend of the corresponding amplitude value and the stress value, calculate the absolute value of the difference between the change trends of the two, and obtain the average value of the sum of the absolute values ​​of the difference of all change trends within a period of time as the change credibility at the target moment.

[0014] Preferably, the deviation degree of the stress value includes:

[0015] Taking any moment as the target moment, the fitting curve and stress fitting value of the stress value at the target moment are obtained, and the sum of the absolute differences between the stress value at the target moment and the stress fitting value, and the ratio of the sum of the absolute differences between the stress value at all historical moments and the stress fitting value, are taken as the deviation degree of the stress value at the target moment.

[0016] The effect is that through continuous monitoring and analysis of stress values, the reliability of the power system can be improved, power outages caused by spacer failure can be reduced, and the deviation of actual stress values ​​of spacers from predicted values ​​can be accurately identified, thereby optimizing the allocation of maintenance resources and giving priority to spacers with higher risks.

[0017] Preferably, the step of obtaining the best split point of the current node by using the priority includes:

[0018] Obtain the priority of the split point at each moment, randomly select stress data as the sub-dataset for building the isolation tree based on the stress data at any moment and all previous moments, select the moment with the highest priority of the split points at different moments in the sub-dataset as the best split point for the current node, and divide the sub-dataset into two sub-nodes; recursively complete the above operations for the two sub-nodes until all data are separately split out to complete the construction of the isolation tree.

[0019] The effect is that by calculating the priority at each moment, the data point at a certain moment can be quickly identified as the best split point, which helps to build a more balanced and efficient isolation tree to ensure that the depth of the tree is not too large, thereby avoiding overfitting, while maintaining the diversity of the tree.

[0020] Preferably, obtaining the abnormal stress data includes:

[0021] Calculate the path length from the root node to the isolated node in the isolation forest for each data point, obtain the average path length of all trees in the isolation forest for each data point, calculate the average path length of the tree based on the number of samples in the isolation forest to standardize the path length, and calculate the ratio between the two average path lengths, and use the negative exponential power of 2 to obtain the anomaly score of each data point;

[0022] An abnormal score threshold is set, and data points whose abnormal scores are greater than the abnormal score threshold are marked as abnormal stress data.

[0023] Preferably, obtaining the vibration tolerance of the spacer rod includes:

[0024] Obtain all abnormal data points, and for each abnormal data point, calculate the total deviation between the stress value of each abnormal data point and the maximum stress value that can be adjusted by the spacer bar;

[0025] The product of the ratio of abnormal data points to the total number of data points and the total deviation is normalized, and the normalized result is converted into a percentage as the tolerance capacity of the spacer rod vibration.

[0026] The effect is that by calculating the sum of the deviations between the stress values ​​of abnormal data points and the maximum stress value, abnormal stress conditions that are close to or exceed the design limit can be identified, thereby discovering potential risk points in a timely manner.

[0027] Preferably, judging the use status of the spacer rod according to the tolerance capacity and a preset tolerance threshold value includes:

[0028] In response to the vibration tolerance of the spacer bar being continuously lower than a preset threshold, it is necessary to consider replacing the spacer bar with a new one to ensure the safety and stability of the transmission line.

[0029] The present invention has the following effects:

[0030] 1. The method of obtaining the optimal segmentation point of the current node through priority in the present invention can not only improve the segmentation efficiency and accuracy of the model, but also optimize resource allocation, enhance the generalization ability of the model, improve the interpretability of the model, and improve the real-time performance of anomaly detection.

[0031] 2. The present invention can reduce the amount of calculation, improve the abnormality detection rate, simplify the model, focus on key indicators, improve the sensitivity of detection, and use more reliable data to effectively evaluate the vibration tolerance of the spacer rod by only constructing an isolated tree of stress data. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0033] Figure 1 It is a method flow chart of steps S1 to S4 in a method for detecting vibration tolerance of a spacer rod according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0035] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0036] Application scenario: The spacer is used to maintain the relative position and spacing between the conductors. When the conductors swing differently, the stress is changed in real time to restore the conductors to a stable state. In this process, the stress at the connection between the spacer and the conductor changes with the actual vibration amplitude of the transmission line, and the change trend is consistent, so as to evaluate the vibration tolerance of the spacer.

[0037] Reference Figure 1 A method for detecting the vibration tolerance of a spacer rod includes steps S1 to S4, which are specifically as follows:

[0038] S1: Based on the time series, the amplitude value of the conductor and the stress value of the spacer bar corresponding to the conductor connection are obtained to obtain the credibility of the change between the amplitude value and the stress value at each moment in history.

[0039] Install the stress sensor at the key position where the spacer bar and the conductor are connected (to ensure that the sensor can accurately capture the stress changes in this area) to monitor the stress value in real time; fix the displacement sensor on the conductor with a professional clamp to monitor the amplitude value in real time; obtain the maximum adjustable stress value of the spacer bar; in the embodiment, all sensors are preset to collect data every 5 seconds.

[0040] The slope difference between the amplitude value and stress value data at each moment in history and the previous moment is calculated to obtain the change trend of the historical amplitude value and the change trend of the corresponding stress value;

[0041] The changing trends of the historical amplitude values ​​and the corresponding stress values ​​are standardized respectively, and the standardized differences between the two at all historical moments are summed up. The output range of the summed data is adjusted using the inverse tangent function, and normalized to obtain the credibility of the change at the current moment.

[0042] Specifically, the change credibility satisfies the following relationship:

[0043] ;

[0044] In the formula, Indicates Credibility changes all the time. Indicated in The first moment in history before the moment The changing trend of the amplitude at each moment, Indicated in The first moment in history before the moment The stress variation trend at each moment is Indicates The standard deviation of the change trend of the amplitude of all historical moments before the moment, Indicates The standard deviation of the stress change trend at all historical moments before the moment, Indicates The index values ​​of all historical moments before the moment (excluding the time), Indicates The number of all historical moments before the moment (excluding the time), Indicates The exponential function with base , Represents the inverse tangent function.

[0045] That is to say, In order to limit the value to The smaller the standardized difference at each historical moment, the more synchronized the changes in amplitude and stress at each historical moment are, which means that the performance of the spacer at the historical moment is reliable. The more reliable the change in the moment data is, the better.

[0046] Specifically, the value range of the inverse tangent function is , where the result of the inverse tangent function is multiplied by , scale and negate; add the scaled arctangent value to , adjust its range to make it more suitable for the input of the exponential function.

[0047] Further explanation: the closer the change credibility is to 1, the The more synchronized the amplitude and stress change trends at the moment, the more stable and reliable the performance of the spacer is; the closer the change credibility is to 0, the better the performance of the spacer is. The more out of sync the amplitude and stress change trends at each moment are, the more likely there is a problem with the performance of the spacer.

[0048] In addition, in another embodiment, the two sets of data are processed separately by using the exponential moving average method (EMA), and then the credibility of the change is evaluated by combining their change trends. The specific steps are as follows:

[0049] Calculate the amplitude and stress values ​​separately :

[0050] For the amplitude value and stress values time series data, and calculate their .

[0051] Set the smoothing factor for the amplitude value and smoothing factor of stress values ,generally , where is the total number of observations.

[0052] initial The value can be the first observation of the time series or a simple moving average of the first few observations. ;

[0053] For the amplitude value, , where Indicates Amplitude value at time value, represents the smoothing factor of the amplitude value, Indicates The amplitude value at the moment, Indicates Amplitude value at time value;

[0054] For stress values, In the formula, Indicates The stress value at the moment value, represents the smoothing factor of the stress value, Indicates The stress value at time, Indicates The stress value at the moment value;

[0055] Calculate the difference in trend: , where Indicates The amplitude value at the moment Value and stress value The difference between the values, Indicates Amplitude value at time value, Indicates The stress value at the moment value;

[0056] Calculate the change credibility: , where Indicates Credibility changes all the time. Indicates time period The number of data in Indicates time period Neidi data, Indicates The amplitude value at the moment Value and stress value The difference between the values.

[0057] It is further explained that a smaller change credibility indicates that the change trends of the amplitude and stress value are relatively consistent, and the change credibility is higher; a larger change credibility indicates that the change trends of the amplitude and stress value are inconsistent, and the change credibility is lower.

[0058] S2: Fitting curves to the data points of stress values ​​at all times according to the least squares method to calculate the degree of deviation of stress values ​​at all times.

[0059] Taking any moment as the target moment, the fitting curve and stress fitting value of the stress value at the target moment are obtained, and the sum of the absolute differences between the stress value at the target moment and the stress fitting value, and the ratio of the sum of the absolute differences between the stress value at all historical moments and the stress fitting value, are taken as the deviation degree of the stress value at the target moment.

[0060] Specifically, the degree of deviation satisfies the following relationship:

[0061] ;

[0062] In the formula, Indicates The degree of deviation of momentary stress, Indicates Time stress, Indicates The stress fitting value at the moment, Indicates The first moment in history before the moment The stress at that moment, Indicates The first moment in history before the moment The stress fitting value at the moment, Indicates The index values ​​of all historical moments before the moment (excluding the time), Indicates The number of all historical moments before the moment (excluding the time).

[0063] To further explain, Indicates The absolute difference between the stress at the moment and the corresponding fitting value. The larger the value, the There is a significant deviation between the stress at the moment and the fitted value; Indicates The sum of the absolute differences between the stress at all historical moments before the moment and the fitted value. The smaller the value, the better the stress at the moment. The deviation between the stress at all moments before the moment and the fitting value is small, indicating that the fitting effect is relatively good; Indicates The greater the value, the better the fitting effect. There is a significant deviation between the stress at the moment and the fitted value, which means that The more the stress at a moment deviates from the overall trend, the higher the deviation of the stress at the current moment is considered.

[0064] S3: Use the isolation forest algorithm to obtain the priority of the split point according to the credibility and deviation of the changes at each moment, use the priority to obtain the best split point of the current node, and recursively complete the construction of the isolation tree.

[0065] Specifically, the priority satisfies the following relationship:

[0066] ;

[0067] In the formula, Indicates The priority of the time division point, Indicates Credibility changes all the time. Indicates The degree of deviation of the stress value at the moment, Indicates An exponential function with base .

[0068] That is to say, The greater the priority of the time segmentation point, the more reliable the performance of the spacer bar is, which means that the stress data fitting trend result is more accurate. The higher the deviation of the stress value at the moment, the more problems there are in the stress data at the current moment. This further shows that in the scenario of isolated trees, using data points that may have problems as split points will confuse the model between normal and abnormal, reducing the accuracy of abnormality detection. Therefore, if there are problems with the stress data at the current moment, the priority of the current moment as a split point is considered to be lower.

[0069] Obtain the priority of the split point at each moment, randomly select stress data as the sub-dataset for building the isolation tree based on the stress data at any moment and all previous moments, select the moment with the highest priority of the split points at different moments in the sub-dataset as the best split point for the current node, and divide the sub-dataset into two sub-nodes; recursively complete the above operations for the two sub-nodes until all data are separately split out to complete the construction of the isolation tree.

[0070] By constructing an isolation tree only for stress values, stress data is usually smaller than a data set containing vibration values. By focusing on stress data, anomalies directly related to the vibration tolerance of the spacer can be captured more accurately; at the same time, stress values ​​are key indicators for measuring the performance of spacer bars, so monitoring the abnormal changes in these data points can more directly evaluate the health of the spacer bars. In summary, constructing an isolation tree only for stress data can effectively evaluate the vibration tolerance of spacer bars by reducing the amount of calculation, improving the anomaly detection rate, simplifying the model, focusing on key indicators, improving the sensitivity of monitoring, and using more reliable data.

[0071] S4: According to the constructed isolated forest, abnormal stress data is obtained, and the tolerance of the spacer bar vibration is calculated according to the proportion of abnormal stress data points and the sum of the deviations between the stress values ​​of all abnormal stress data points and the maximum adjustable stress value of the spacer bar. The use status of the spacer bar is determined according to the tolerance and the preset tolerance threshold.

[0072] Calculate the path length from the root node to the isolated node in the isolation forest for each data point, obtain the average path length of all trees in the isolation forest for each data point, calculate the average path length of the tree based on the number of samples in the isolation forest to standardize the path length, and calculate the ratio between the two average path lengths, and use the negative exponential power of 2 to obtain the anomaly score of each data point;

[0073] Set an anomaly score threshold and mark data points with an anomaly score greater than the anomaly score threshold as abnormal stress data.

[0074] Specifically, the abnormal stress data satisfies the following relationship:

[0075] ;

[0076] In the formula, Represents sample data in isolation forest Middle The anomaly score of the data point, Indicates The path length of a data point from the root node to the leaf node, Indicates The average path length of all trees in the isolation forest for a data point, represents the average path length of the tree, Represents sample data.

[0077] In this embodiment, the anomaly score threshold is 0.8, which can be adjusted according to actual conditions.

[0078] Obtain all abnormal data points, and for each abnormal data point, calculate the total deviation between the stress value of each abnormal data point and the maximum stress value that can be adjusted by the spacer bar;

[0079] The product of the ratio of abnormal data points to the total number of data points and the total deviation is normalized, and the normalized result is converted into a percentage as the tolerance capacity of the spacer rod vibration.

[0080] Specifically, the tolerance of the spacer vibration is obtained to satisfy the following relationship:

[0081] ;

[0082] In the formula, Indicates the vibration tolerance of the spacer rod. Indicates The total number of data points at the current moment and all previous historical moments, Represents the number of all abnormal data points, Indicates that the spacer rod can adjust the maximum stress value. Indicates The stress value of the abnormal data point, Indicates An exponential function with base .

[0083] That is to say, It indicates the proportion of abnormal data points to the total data points. The larger the value, the more abnormal stress conditions the spacer bar encounters during operation, that is, the more data points that do not meet the expected stress changes. In order to avoid the sum result being 0; the smaller the total deviation value, the closer the abnormal stress value is to the maximum value of the spacer bar, which means that the abnormal stress borne by the spacer bar is closer to its design limit, resulting in the spacer bar being unable to effectively stabilize the conductor; if the spacer bar has more data points that do not meet the expected stress changes in historical work and the corresponding stress values ​​of these abnormal data points are closer to the stress limit value it bears, it is considered that the vibration tolerance of the spacer bar is worse.

[0084] In response to the vibration tolerance of the spacer bar being continuously lower than a preset threshold, it is necessary to consider replacing the spacer bar with a new one to ensure the safety and stability of the transmission line.

[0085] In this embodiment, the preset threshold is 20%, which can be adjusted according to actual conditions.

[0086] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0087] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for detecting the vibration tolerance of a spacer rod, characterized in that: include: Based on the time series, the amplitude value of the conductor and the stress value of the spacer bar corresponding to the conductor connection are obtained in history, and the credibility of the change between the amplitude value and the stress value at each moment in history is obtained; A curve is fitted to the data points of stress values ​​at all times according to the least square method to calculate the degree of deviation of stress values ​​at all times; Using the isolation forest algorithm, the priority of the segmentation point is obtained according to the degree of deviation of the credibility of the change at each moment, and the priority of the segmentation point at each moment is obtained. According to the stress data at any moment and all previous moments, the stress data is randomly selected as the sub-dataset for building the isolation tree, and the moment with the highest priority of the segmentation points at different moments in the sub-dataset is selected as the best segmentation point for the current node, and the sub-dataset is divided into two sub-nodes; the above operations are recursively completed for the two sub-nodes until all the data are separately segmented to complete the construction of the isolation tree; According to the constructed isolation forest, abnormal stress data is obtained, all abnormal data points are obtained, and for each abnormal data point, the total deviation between the stress value of each abnormal data point and the maximum stress value that can be adjusted by the spacer bar is calculated; the ratio of the abnormal data points to the total number of data points and the product of the total deviation are normalized, and the normalized result is converted into a percentage as the tolerance of the spacer bar vibration, and the use status of the spacer bar is judged according to the tolerance and the preset tolerance threshold; The priority satisfies the following relationship: , where Indicates The priority of the time division point, Indicates Credibility changes all the time. Indicates The degree of deviation of the stress value at the moment, Indicates An exponential function with base .

2. A method for detecting the vibration tolerance of a spacer rod according to claim 1, characterized in that: The change credibility includes: The slope difference between the amplitude value and stress value data at each moment in history and the previous moment is calculated to obtain the change trend of the historical amplitude value and the change trend of the corresponding stress value; The changing trends of the historical amplitude values ​​and the corresponding stress values ​​are standardized respectively, and the standardized differences between the two at all historical moments are summed up. The output range of the summed data is adjusted using the inverse tangent function, and normalized to obtain the credibility of the change at the current moment.

3. A method for detecting the vibration tolerance of a spacer rod according to claim 1, characterized in that: The change credibility also includes: Take the current time as the target time and get the time before the target time The amplitude and stress values ​​within the target time are calculated by the exponential moving average method. The amplitude and stress values ​​within , set the smoothing factors of the amplitude value and the stress value respectively, obtain the change trend of the corresponding amplitude value and the stress value, calculate the absolute value of the difference between the change trends of the two, and obtain the average value of the sum of the absolute values ​​of the difference of all change trends within a period of time as the change credibility at the target moment.

4. A method for detecting the vibration tolerance of a spacer rod according to claim 1, characterized in that: The degree of deviation of the stress value includes: Taking any moment as the target moment, the fitting curve and stress fitting value of the stress value at the target moment are obtained, and the sum of the absolute differences between the stress value at the target moment and the stress fitting value, and the ratio of the sum of the absolute differences between the stress value at all historical moments and the stress fitting value, are taken as the deviation degree of the stress value at the target moment.

5. A method for detecting the vibration tolerance of a spacer rod according to claim 1, characterized in that: Acquiring the abnormal stress data includes: Calculate the path length from the root node to the isolated node in the isolation forest for each data point, obtain the average path length of all trees in the isolation forest for each data point, calculate the average path length of the tree based on the number of samples in the isolation forest to standardize the path length, and calculate the ratio between the two average path lengths, and use the negative exponential power of 2 to obtain the anomaly score of each data point; An abnormal score threshold is set, and data points whose abnormal scores are greater than the abnormal score threshold are marked as abnormal stress data.

6. A method for detecting the vibration tolerance of a spacer rod according to claim 1, characterized in that: The method of judging the use status of the spacer rod according to the tolerance capacity and the preset tolerance threshold value includes: In response to the vibration tolerance of the spacer bar being continuously lower than a preset threshold, it is necessary to consider replacing the spacer bar with a new one to ensure the safety and stability of the transmission line.

Images