Dynamic monitoring method of transmission tower disaster based on stress analysis

Abstract

The present invention relates to the technical field of transmission tower disaster monitoring, and specifically to a method for dynamic monitoring of transmission tower disasters based on stress analysis, comprising: obtaining vibration data of the top of the transmission tower and first vibration data of each conductor connected to the transmission tower, obtaining several environmental parameter data of the transmission tower; obtaining the first influence degree of each environmental parameter data on the vibration data; obtaining the predicted strain value of the top of the transmission tower at a future moment according to the vibration data, the first influence degree and the change in the environmental parameter value; and performing dynamic monitoring of the transmission tower disaster according to the predicted strain value. The present invention can timely and effectively carry out disaster warning, and improves the accuracy of transmission tower disaster detection.

Description

Technical Field

[0001] The invention relates to the technical field of transmission iron tower disaster monitoring, and in particular to a transmission iron tower disaster dynamic monitoring method based on stress analysis. Background Art

[0002] As an important infrastructure for power transmission, transmission towers bear the key tasks of supporting high-voltage cables and maintaining the stable operation of the power grid. In the long-term operation, the tower structure needs to withstand the influence of external natural factors such as wind load, ice and snow, temperature changes and earthquakes. It is easy to cause structural failure or collapse due to stress accumulation or sudden disasters. This will not only cause huge economic losses, but also may cause large-scale power outages and affect the normal operation of society.

[0003] Traditional transmission tower monitoring mainly relies on manual inspections and limited sensor equipment, which makes it difficult to capture the stress changes of the tower in a timely manner, especially under extreme weather conditions or in large-scale power networks, and the monitoring efficiency is low. In addition, it is difficult to fully reflect the health status of the overall structure of the tower based only on the status data of a single stress detection node. In the actual power grid, the mechanical state and stress distribution of the wires connected to the tower are strongly correlated with it, but the existing monitoring methods usually analyze a single tower independently and fail to use the correlation data between towers for collaborative analysis, thereby reducing the accuracy of disaster detection. The existing monitoring methods cannot provide effective disaster warnings, which greatly limits the intelligent and precise development of the monitoring system. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method for dynamic monitoring of transmission tower disasters based on stress analysis.

[0005] The method for dynamic monitoring of transmission tower disaster based on stress analysis of the present invention adopts the following technical scheme:

[0006] An embodiment of the present invention provides a method for dynamic monitoring of transmission tower disasters based on stress analysis, the method comprising the following steps:

[0007] Acquire vibration data of the top of the transmission tower and first vibration data of each conductor connected to the transmission tower, and acquire data of several environmental parameters of the transmission tower;

[0008] Obtaining a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all the first vibration data;

[0009] According to the environmental parameter data, an environmental vector corresponding to each environmental parameter data at each moment is obtained, and the environmental vector is clustered to obtain a plurality of clusters corresponding to each environmental parameter data; according to the clusters corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector, a first influence degree of each environmental parameter data on the vibration data is obtained;

[0010] Obtaining a preset neighborhood data segment of the environmental parameter value at each moment in each environmental parameter data; obtaining the similarity of environmental changes at any two moments according to the correlation between different preset neighborhood data segments;

[0011] According to the similarity and the change of environmental parameter values, the change of environmental parameter values ​​of each environmental parameter data at each moment is obtained; according to the vibration data, the first impact degree and the change of environmental parameter values, the predicted strain value of the top of the transmission tower at the future moment is obtained; according to the predicted strain value, the transmission tower is dynamically monitored for disasters.

[0012] Furthermore, the step of obtaining a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all the first vibration data includes the following specific steps:

[0013] Arrange all vibration amplitudes in the vibration data of the top of the transmission tower in columns to obtain a column vector of the vibration data; obtain a column vector of each first vibration data according to a method for obtaining a column vector of the vibration data; arrange the column vector of the vibration data and the column vectors of all the first vibration data in rows to obtain a vibration data matrix;

[0014] Factor analysis is performed on the vibration data matrix to obtain a number of vibration factor vectors and the first coefficient of each vibration factor vector.

[0015] Furthermore, the step of obtaining the environment vector corresponding to each environment parameter data at each moment according to the environment parameter data, clustering the environment vectors, and obtaining a plurality of clusters corresponding to each environment parameter data includes the following specific steps:

[0016] All environmental parameter data of the transmission tower are The vector of environmental parameter values ​​at the moment is recorded as the initial environmental vector. The environmental parameter value of the environmental parameter data is removed to obtain the The environmental parameter data is in The environment vector corresponding to the moment The environmental vectors corresponding to the environmental parameter data at all times are clustered by DBSCAN, and the distance measurement uses the Euclidean distance between environmental vectors to obtain the first Several clusters corresponding to environmental parameter data.

[0017] Furthermore, the first influence degree of each environmental parameter data on the vibration data is obtained according to the cluster corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector, and the specific steps include the following:

[0018]

[0019] in, Indicates The number of clusters corresponding to each environmental parameter data; Indicates The environmental parameter data corresponds to The number of moments corresponding to the environment vector in each cluster; Indicates The vibration factor vector The first-order backward difference of elements; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Represents a hyperparameter that prevents the denominator from being 0; Indicates Environmental parameter data for the The influence degree of each vibration factor vector;

[0020] According to the first coefficient and the influence degree of the vibration factor vector, the first influence degree of each environmental parameter data on the vibration data is obtained.

[0021] Furthermore, the first influence degree of each environmental parameter data on the vibration data is obtained according to the first coefficient and influence degree of the vibration factor vector, and the specific steps include the following:

[0022]

[0023] in, Represents the number of vibration factor vectors; Indicates The first coefficient of the vibration factor vector; Indicates The first degree of influence of environmental parameter data on vibration data.

[0024] Furthermore, the similarity of environmental changes at any two moments is obtained according to the correlation between different preset neighborhood data segments, and the specific steps include the following:

[0025]

[0026] in, Indicates the number of environmental parameter data; Indicates Environmental parameter data The preset neighborhood data segment of the environmental parameter value at the moment The Pearson correlation coefficient of the preset neighborhood data segment of the environmental parameter value at each moment; is a linear normalization function; Indicates taking the absolute value; Indicates The moment and The similarity of environmental changes at each moment.

[0027] Furthermore, the step of obtaining the environmental parameter value change of each environmental parameter data at each moment according to the similarity and the change of the environmental parameter value includes the following specific steps:

[0028] The first The time when the last environmental parameter value in the environmental parameter data is located is recorded as moment; The moment before the moment, recorded as a moment;

[0029]

[0030] in, Indicates The moment and The similarity of environmental changes at each moment; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Indicates The environmental parameter data is in The change in the environmental parameter value at a certain moment.

[0031] Furthermore, the step of obtaining the predicted strain value of the top of the transmission tower at a future time according to the vibration data, the first impact degree and the change in the environmental parameter value includes the following specific steps:

[0032]

[0033] in, Indicates the number of environmental parameter data; Indicates The first degree of influence of environmental parameter data on vibration data; Indicates The environmental parameter data is in The change of environmental parameter value at each moment; Indicates the vibration data of the top of the transmission tower. A vibration amplitude; The vibration data of the top of the transmission tower is shown in The predicted vibration amplitude at each moment;

[0034] According to the predicted vibration amplitude, the predicted strain value of the top of the transmission tower at the future moment is obtained.

[0035] Furthermore, the step of obtaining the predicted strain value of the top of the transmission tower at a future time according to the predicted vibration amplitude includes the following specific steps:

[0036] Obtaining vibration data on the top of transmission towers in the future Several predicted vibration amplitudes at the moment, The vibration data of the top of the transmission tower will be used as the preset hyperparameter in the future. Several predicted vibration amplitudes at the moment are used as input and input into the LSTM neural network. The loss function selects the cross entropy loss function to perform network training and output the top of the transmission tower in the future. Several predicted strain values ​​at a certain moment.

[0037] Furthermore, the disaster dynamic monitoring of the transmission tower according to the predicted strain value includes the following specific steps:

[0038] Preset catastrophic stress threshold, if the top of the transmission tower Among the predicted strain values ​​at a certain moment, if there is a predicted strain value greater than the preset catastrophic stress threshold, the transmission tower will be catastrophic; otherwise, the transmission tower will not be catastrophic.

[0039] The beneficial effects of the technical solution of the present invention are as follows: the present invention predicts the stress state of the iron tower in real time by combining the vibration data of the top of the transmission iron tower, the first vibration data of each wire connected to the transmission iron tower and the weather influence factors, thereby improving the accuracy and prediction ability of disaster detection, so as to realize intelligent and dynamic tower disaster monitoring. When determining the first influence degree of each environmental parameter data on the vibration data, the first influence degree is determined by analyzing the cluster corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector, accurately describing the influence of the environment on the vibration data of the top of the transmission iron tower, and improving the accuracy, scientificity and objectivity of the first influence degree. When determining the predicted strain value of the top of the transmission iron tower at a future moment, the predicted strain value of the top of the transmission iron tower at a future moment is determined by analyzing the similarity of environmental changes at any two moments, and combining the change amount of the environmental parameter value and the first influence degree. In the calculation process, the influence of the environment on the transmission iron tower is fully considered, and the accuracy, scientificity and objectivity of the predicted strain value are improved. Finally, the transmission iron tower is dynamically monitored for disasters according to the predicted strain value. It solves the problem that traditional transmission tower monitoring mainly relies on manual inspections and limited sensor equipment, making it difficult to capture the stress changes of the tower in a timely manner, especially in extreme weather conditions or in large-scale power networks, and the monitoring efficiency is low. It can provide timely and effective disaster warnings and improve the accuracy of transmission tower disaster detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flowchart of the steps of a method for dynamic monitoring of transmission tower disasters based on stress analysis provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of the transmission tower disaster dynamic monitoring method based on stress analysis proposed by the present invention are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of the method for dynamic monitoring of transmission tower disaster based on stress analysis provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of a method for dynamic monitoring of transmission tower disasters based on stress analysis provided by an embodiment of the present invention, the method comprising the following steps:

[0046] Step S001, obtaining vibration data of the top of a transmission tower and first vibration data of each conductor connected to the transmission tower, and obtaining data of several environmental parameters of the transmission tower.

[0047] It should be noted that the dynamic monitoring of transmission tower disasters usually monitors the vibration data of the top of the transmission tower. The main purpose of this embodiment is to obtain the predicted strain value of the top of the transmission tower at a future time, and then perform dynamic monitoring of the transmission tower disasters based on the predicted strain value. Before starting the analysis, first collect data related to the transmission tower.

[0048] Specifically, the vibration data of the top of the transmission tower and the first vibration data of each wire connected to the transmission tower are obtained, and several environmental parameter data of the transmission tower are obtained, as follows:

[0049] The stress changes of key structural parts of the tower are detected by surface strain gauges with a detection accuracy of ±0.5~1με. The main detection points are the welding or bolt nodes connecting the main column and the cross arm, and the measurement unit is microstrain ( ); The following is the strain safety range at some test locations: The strain range at the bottom of the main column is generally The main column is mainly subjected to longitudinal compressive stress, and the strain range is within the elastic range of the material; the strain range of the cross arm suspension point is generally within Due to the tension of the wire, the strain changes greatly.

[0050] The inclination accelerometer is used to measure the vibration amplitude and frequency at the top of the tower (unit: ), reflecting the vibration or impact intensity. And because the state of the wires between the transmission towers has a great influence on the strain of the towers on both sides, it is considered to install acceleration sensors on the wires to detect the vibration data of the wires in real time, and obtain the vibration data of the top of the transmission tower and the first vibration data of each wire connected to the transmission tower.

[0051] The acceleration range of the above sensors is: , (Common wind vibration frequency is ).

[0052] The six meteorological elements sensors mainly detect environmental parameters such as ambient temperature, humidity, wind speed, wind direction, atmospheric pressure and rainfall, and obtain the ambient temperature data, ambient humidity data, ambient wind speed data, ambient wind direction data, ambient atmospheric pressure data and ambient rainfall data of the transmission tower. The accuracy range of each parameter is as follows:

[0053] temperature: ; Accuracy .

[0054] humidity: ; Accuracy .

[0055] Wind speed: ; Accuracy .

[0056] wind direction: ; Accuracy .

[0057] Atmospheric Pressure: hPa; accuracy hPa.

[0058] Rainfall: mm / h; accuracy mm.

[0059] It should be noted that the time series length and sampling frequency of the vibration data at the top of the transmission tower, the first vibration data of each wire connected to the transmission tower, and several environmental parameter data are the same. The vibration data and the first vibration data both contain multiple moments, each moment corresponds to a vibration amplitude; the environmental parameter data contains multiple moments, each moment corresponds to an environmental parameter value.

[0060] At this point, the vibration data of the top of the transmission tower and the first vibration data of each conductor connected to the transmission tower are obtained, and several environmental parameter data of the transmission tower are obtained.

[0061] Step S002: Obtain a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all first vibration data.

[0062] It should be noted that the tower and the conductor are linked structures. Monitoring only the tower or the conductor may miss key vibration information, fail to accurately assess the stability of the system, and lead to low accuracy of subsequent predictions. By performing factor analysis on the vibration data of the tower and the conductor connected to it, several major vibration influencing factors can be extracted from the complex and multi-dimensional vibration data, thereby systematically revealing the vibration mode of the tower and the conductor and their interaction relationship.

[0063] Specifically, obtaining a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all the first vibration data includes the following steps:

[0064] First, a vibration data matrix is ​​obtained according to the vibration data and all the first vibration data.

[0065] Secondly, factor analysis is performed on the vibration data matrix to obtain a number of vibration factor vectors and the first coefficient of each vibration factor vector.

[0066] Specifically, a vibration data matrix is ​​obtained according to the vibration data and all the first vibration data, as follows:

[0067] Arrange all vibration amplitudes in the vibration data at the top of the transmission tower in columns to obtain a column vector of the vibration data; obtain a column vector of each first vibration data according to a method for obtaining a column vector of the vibration data; arrange the column vector of the vibration data and the column vectors of all the first vibration data in rows to obtain a vibration data matrix.

[0068] Further, factor analysis is performed on the vibration data matrix to obtain a number of vibration factor vectors and the first coefficient of each vibration factor vector. It should be noted that factor analysis is performed on the vibration data matrix to obtain a number of vibration factor vectors and the first coefficient of each vibration factor vector is an existing method of factor analysis, which will not be described in detail in this embodiment.

[0069] It should be noted that after the relationship between each original vibration data and multiple factors is obtained, a subsequent analysis is performed to assist in analyzing the impact of weather on the vibration data.

[0070] At this point, several vibration factor vectors and the first coefficient of each vibration factor vector are obtained.

[0071] Step S003, obtaining the environmental vector corresponding to each environmental parameter data at each moment according to the environmental parameter data, clustering the environmental vectors to obtain a number of clusters corresponding to each environmental parameter data; obtaining the first influence degree of each environmental parameter data on the vibration data according to the clusters corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector.

[0072] It should be noted that after obtaining each factor, the impact of environmental factors on tower vibration is considered, because the impact of environmental factors on tower vibration is complex and indirect. For example, wind will directly act on the tower to produce vibration, and will also indirectly affect the tower vibration by acting on the wire to produce tension changes. Therefore, through factor analysis, these complex direct and indirect relationships can be decoupled, and the effect of the environment on each vibration factor can be clearly quantified, which is convenient for comprehensive evaluation of the overall impact of the environment on tower vibration. First, consider the influence coefficient of the environment on a single vibration factor. Since different environmental factors have different effects on vibration data, consider analyzing it by controlling variables. For example, when analyzing the impact of a certain dimension of environmental data on any vibration factor, consider clustering analysis of the remaining multiple dimensions of environmental data.

[0073] Specifically, the environment vector corresponding to each environment parameter data at each moment is obtained according to the environment parameter data, and the environment vector is clustered to obtain several clusters corresponding to each environment parameter data, as follows:

[0074] All environmental parameter data of the transmission tower are The vector of environmental parameter values ​​at the moment is recorded as the initial environmental vector. The environmental parameter value of the environmental parameter data is removed to obtain the The environmental parameter data is in The environment vector corresponding to the moment The environmental vectors corresponding to the environmental parameter data at all times are clustered by DBSCAN, and the distance measurement uses the Euclidean distance between environmental vectors to obtain the first Several clusters corresponding to environmental parameter data.

[0075] It should be noted that the neighborhood radius and the minimum number of samples of the DBSCAN clustering algorithm are 2 and 5 respectively in this embodiment.

[0076] It should be noted that within a single cluster, due to the high similarity between the remaining dimensions, the impact of data changes in a single dimension on a single vibration factor is more uniform.

[0077] Specifically, according to the cluster corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector, the first influence degree of each environmental parameter data on the vibration data is obtained, which includes the following steps:

[0078] Firstly, according to the clusters corresponding to the environmental parameter data, the environmental parameter data and the vibration factor vector, the influence degree of each environmental parameter data on each vibration factor vector is obtained.

[0079] Secondly, according to the first coefficient and the influence degree of the vibration factor vector, the first influence degree of each environmental parameter data on the vibration data is obtained.

[0080] Specifically, according to the clusters corresponding to the environmental parameter data, the environmental parameter data and the vibration factor vector, the influence degree of each environmental parameter data on each vibration factor vector is obtained, as follows:

[0081]

[0082] in, Indicates The number of clusters corresponding to each environmental parameter data; Indicates The environmental parameter data corresponds to The number of moments corresponding to the environment vector in each cluster; Indicates The vibration factor vector The first-order backward difference of elements; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Represents a hyperparameter that prevents the denominator from being 0. In this embodiment, to give a narrative; Indicates Environmental parameter data for the The influence degree of a vibration factor vector.

[0083] It should be noted that after obtaining the influence of each dimension of environmental data on each vibration factor, when analyzing the influence of each environmental factor on the vibration data, the influence of each vibration factor is comprehensively considered through the above-mentioned factor coefficients.

[0084] Specifically, according to the first coefficient and the influence degree of the vibration factor vector, the first influence degree of each environmental parameter data on the vibration data is obtained, as follows:

[0085]

[0086] in, Represents the number of vibration factor vectors; Indicates The first coefficient of the vibration factor vector; Indicates Environmental parameter data for the The influence degree of each vibration factor vector; Indicates The first degree of influence of environmental parameter data on vibration data.

[0087] It should be noted that Environmental parameter data for the The greater the influence of the vibration factor vector, the greater the The larger the first coefficient of a vibration factor vector is, the greater the impact on the vibration data is.

[0088] At this point, the first influence degree of each environmental parameter data on the vibration data is obtained.

[0089] Step S004: obtaining a preset neighborhood data segment of the environmental parameter value at each moment in each environmental parameter data; and obtaining the similarity of environmental changes at any two moments according to the correlation between different preset neighborhood data segments.

[0090] It should be noted that vibration is the external manifestation of the change in the stress state of the tower structure, and its frequency and amplitude directly reflect the dynamic stress characteristics of the tower. Vibration data is highly correlated with strain data. For example, areas with larger vibration amplitudes are often accompanied by more significant stress changes. Therefore, vibration data analysis can provide a clear mechanical basis for subsequent strain analysis. Therefore, based on the prediction results of vibration data, it is considered to perform predictive analysis on strain data.

[0091] It should be noted that the vibration data at the current moment is predicted based on the environmental data at the historical moment. Considering that both environmental factors and vibration data in the historical data have certain uncertainties and randomness, the prediction of vibration data at a single time point may be affected by accidental factors, resulting in large deviations. The similarity is calculated using the Pearson correlation coefficient, which can effectively screen the data most relevant to the current environment. By comparing the similarity between environmental data in different time windows in the historical data and the current environmental data, the scenario most similar to the current situation can be found from the historical data, thereby improving the accuracy of subsequent predictions.

[0092] Specifically, the preset neighborhood data segment of the environmental parameter value at each moment in each environmental parameter data is obtained, as follows:

[0093] For Environmental parameter data The environmental parameter value at the moment The environmental parameter values ​​at the moment Before the environmental parameter value at the moment The data segment consisting of the environmental parameter values ​​at each moment is The preset neighborhood data segment of the environmental parameter value at the moment, is the preset hyperparameter. Give a narrative.

[0094] It should be noted that if there are no 20 environmental parameter values ​​before a certain environmental parameter value, the preset neighborhood data segment will not be obtained.

[0095] Furthermore, according to the correlation between different preset neighborhood data segments, the similarity of environmental changes at any two moments is obtained, as follows:

[0096]

[0097] in, Indicates the number of environmental parameter data; Indicates Environmental parameter data The preset neighborhood data segment of the environmental parameter value at the moment The Pearson correlation coefficient of the preset neighborhood data segment of the environmental parameter value at each moment; is a linear normalization function used for normalization processing; Indicates taking the absolute value; Indicates The moment and The similarity of environmental changes at each moment.

[0098] It should be noted that when Environmental parameter data The preset neighborhood data segment of the environmental parameter value at the moment The larger the Pearson correlation coefficient of the preset neighborhood data segment of the environmental parameter value at the moment, the The moment and The similarity of environmental changes at each moment is high. The smaller the time, the closer the two moments are. The moment and The similarity of environmental changes at each moment is also high.

[0099] At this point, the similarity of environmental changes at any two moments is obtained.

[0100] Step S005: obtain the change in the environmental parameter value of each environmental parameter data at each moment according to the similarity and the change in the environmental parameter value; obtain the predicted strain value of the top of the transmission tower at a future moment according to the vibration data, the first impact degree and the change in the environmental parameter value; and perform disaster dynamic monitoring of the transmission tower according to the predicted strain value.

[0101] It should be noted that for similar moments in historical environmental change data, if the similarity between two sequences is high at historical moments, the accuracy of estimating the change in environmental data at the next moment can be higher by using the change in the data in the corresponding sequence at the next moment.

[0102] Specifically, according to the similarity and the change of the environmental parameter value, the change amount of the environmental parameter value of each environmental parameter data at each moment is obtained, as follows:

[0103] The first The time when the last environmental parameter value in the environmental parameter data is located is recorded as moment; The moment before the moment, recorded as a moment.

[0104]

[0105] in, Indicates The moment and The similarity of environmental changes at each moment; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Indicates The environmental parameter data is in The change in the environmental parameter value at a certain moment.

[0106] Specifically, according to the vibration data, the first impact degree and the change amount of the environmental parameter value, the predicted strain value of the top of the transmission tower at a future time is obtained, including the steps of:

[0107] Firstly, according to the vibration data, the first impact degree and the change of the environmental parameter value, the predicted vibration amplitude of the vibration data on the top of the transmission tower at a future time is obtained.

[0108] Secondly, according to the predicted vibration amplitude, the predicted strain value of the top of the transmission tower at the future moment is obtained.

[0109] Further, according to the vibration data, the first impact degree and the change in the environmental parameter value, the predicted vibration amplitude of the vibration data at the top of the transmission tower at a future time is obtained, as follows:

[0110]

[0111] in, Indicates the number of environmental parameter data; Indicates The first degree of influence of environmental parameter data on vibration data; Indicates The environmental parameter data is in The change of environmental parameter value at each moment; Indicates the vibration data of the top of the transmission tower. A vibration amplitude; The vibration data of the top of the transmission tower is shown in The predicted vibration amplitude at a certain moment.

[0112] Furthermore, according to the predicted vibration amplitude, the predicted strain value of the top of the transmission tower at the future time is obtained, as follows:

[0113] Obtaining vibration data on the top of transmission towers in the future Several predicted vibration amplitudes at the moment, is the preset hyperparameter. Narrate and use the vibration data of the top of the transmission tower in the future The predicted vibration amplitudes at the moment are used as input and input into the LSTM neural network (Long Short-Term Memory). The cross entropy loss function is selected as the loss function to train the network and output the top of the transmission tower in the future. Several predicted strain values ​​at a certain moment.

[0114] What needs to be explained is that the top of the transmission tower is obtained through the LSTM neural network in the future The predicted strain values ​​at each moment are based on existing methods and will not be described in detail in this embodiment.

[0115] It should be noted that, according to the prediction results of the transmission tower at the detection point, dynamic monitoring of disasters can be achieved by comparing the predicted strain value of the transmission tower with the normal strain range in real time.

[0116] Specifically, the disaster dynamic monitoring of the transmission tower is carried out according to the predicted strain value, as follows:

[0117] The preset catastrophic stress threshold is Describe, if the top of the transmission tower in the future Among the predicted strain values ​​at a certain moment, if there is a predicted strain value greater than the preset catastrophic stress threshold, the transmission tower will be catastrophic; otherwise, the transmission tower will not be catastrophic.

[0118] It should be noted that when the predicted value is within the normal strain range, it indicates that the tower structure is in a safe state; if the predicted value is close to or exceeds the upper limit of the normal strain range, there may be structural abnormalities, reminding relevant personnel to check for hidden dangers in a timely manner.

[0119] Through the above steps, the transmission tower disaster dynamic monitoring method based on stress analysis is completed.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for dynamic monitoring of transmission tower disasters based on stress analysis, characterized in that: The method comprises the following steps: Obtain vibration data of the top of the transmission tower and first vibration data of each conductor connected to the transmission tower, and obtain several environmental parameter data of the transmission tower; the environmental parameter data includes multiple moments, each moment corresponds to an environmental parameter value; Obtaining a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all the first vibration data; According to the environmental parameter data, an environmental vector corresponding to each environmental parameter data at each moment is obtained, and the environmental vector is clustered to obtain a plurality of clusters corresponding to each environmental parameter data; according to the clusters corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector, a first influence degree of each environmental parameter data on the vibration data is obtained; Obtaining a preset neighborhood data segment of the environmental parameter value at each moment in each environmental parameter data; obtaining the similarity of environmental changes at any two moments according to the correlation between different preset neighborhood data segments; According to the similarity and the change of the environmental parameter value, the change of the environmental parameter value of each environmental parameter data at each moment is obtained; according to the vibration data, the first impact degree and the change of the environmental parameter value, the predicted strain value of the top of the transmission tower at a future moment is obtained; according to the predicted strain value, the disaster dynamic monitoring of the transmission tower is carried out; The method of obtaining the similarity of environmental changes at any two moments based on the correlation between different preset neighborhood data segments includes the following specific steps: in, Indicates the number of environmental parameter data; Indicates Environmental parameter data The preset neighborhood data segment of the environmental parameter value at the moment The Pearson correlation coefficient of the preset neighborhood data segment of the environmental parameter value at each moment; is a linear normalization function; Indicates taking the absolute value; Indicates The moment and The similarity of environmental changes at each moment.

2. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 1 is characterized in that: The step of obtaining a plurality of vibration factor vectors and a first coefficient of each vibration factor vector according to the vibration data and all the first vibration data includes the following specific steps: Arrange all vibration amplitudes in the vibration data of the top of the transmission tower in columns to obtain a column vector of the vibration data; obtain a column vector of each first vibration data according to a method for obtaining a column vector of the vibration data; arrange the column vector of the vibration data and the column vectors of all the first vibration data in rows to obtain a vibration data matrix; Factor analysis is performed on the vibration data matrix to obtain a number of vibration factor vectors and the first coefficient of each vibration factor vector.

3. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 1 is characterized in that: The specific steps of obtaining the environment vector corresponding to each environment parameter data at each moment according to the environment parameter data, clustering the environment vectors, and obtaining a plurality of clusters corresponding to each environment parameter data are as follows: All environmental parameter data of the transmission tower are The vector of environmental parameter values ​​at the moment is recorded as the initial environmental vector. The environmental parameter value of the environmental parameter data is removed to obtain the The environmental parameter data is in The environment vector corresponding to the moment The environmental vectors corresponding to the environmental parameter data at all times are clustered by DBSCAN, and the distance measurement uses the Euclidean distance between environmental vectors to obtain the first Several clusters corresponding to environmental parameter data.

4. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 1 is characterized in that: The step of obtaining the first influence degree of each environmental parameter data on the vibration data according to the cluster corresponding to the environmental parameter data, the environmental parameter data, the vibration factor vector and the first coefficient of the vibration factor vector comprises the following specific steps: in, Indicates The number of clusters corresponding to each environmental parameter data; Indicates The environmental parameter data corresponds to The number of moments corresponding to the environment vector in each cluster; Indicates The vibration factor vector The first-order backward difference of elements; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Represents a hyperparameter that prevents the denominator from being 0; Indicates Environmental parameter data for the The influence degree of each vibration factor vector; According to the first coefficient and the influence degree of the vibration factor vector, the first influence degree of each environmental parameter data on the vibration data is obtained.

5. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 4 is characterized in that: The step of obtaining the first influence degree of each environmental parameter data on the vibration data according to the first coefficient and the influence degree of the vibration factor vector includes the following specific steps: in, Represents the number of vibration factor vectors; Indicates The first coefficient of the vibration factor vector; Indicates The first degree of influence of environmental parameter data on vibration data.

6. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 1 is characterized in that: The specific steps of obtaining the environmental parameter value change amount of each environmental parameter data at each moment according to the similarity and the change of the environmental parameter value are as follows: The first The time when the last environmental parameter value in the environmental parameter data is located is recorded as moment; The moment before the moment, recorded as a moment; in, Indicates The moment and The similarity of environmental changes at each moment; Indicates Environmental parameter data The first-order backward difference of the environmental parameter values; Indicates The environmental parameter data is in The change in the environmental parameter value at a certain moment.

7. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 6 is characterized in that: The method of obtaining the predicted strain value of the top of the transmission tower at a future time according to the vibration data, the first impact degree and the change in the environmental parameter value includes the following specific steps: in, Indicates the number of environmental parameter data; Indicates The first degree of influence of environmental parameter data on vibration data; Indicates The environmental parameter data is in The change of environmental parameter value at each moment; Indicates the vibration data of the top of the transmission tower. A vibration amplitude; The vibration data of the top of the transmission tower is shown in The predicted vibration amplitude at each moment; According to the predicted vibration amplitude, the predicted strain value of the top of the transmission tower at the future moment is obtained.

8. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 7 is characterized in that: The specific steps of obtaining the predicted strain value of the top of the transmission tower at a future time according to the predicted vibration amplitude are as follows: Obtaining vibration data on the top of transmission towers in the future Several predicted vibration amplitudes at the moment, The vibration data of the top of the transmission tower will be used as the preset hyperparameter in the future. Several predicted vibration amplitudes at the moment are used as input and input into the LSTM neural network. The loss function selects the cross entropy loss function to perform network training and output the top of the transmission tower in the future. Several predicted strain values ​​at a certain moment.

9. The method for dynamic monitoring of transmission tower disaster based on stress analysis according to claim 8 is characterized in that: The specific steps of carrying out the disaster dynamic monitoring of the transmission tower according to the predicted strain value are as follows: Preset catastrophic stress threshold, if the top of the transmission tower Among the predicted strain values ​​at a certain moment, if there is a predicted strain value greater than the preset catastrophic stress threshold, the transmission tower will be catastrophic; otherwise, the transmission tower will not be catastrophic.

Images