CN-121997062-A - Power transmission tower state evaluation method and system based on inhaul cable tension

CN121997062ACN 121997062 ACN121997062 ACN 121997062ACN-121997062-A

Abstract

The invention discloses a power transmission tower state evaluation method and system based on cable tension, which are applied to the field of wind vibration control of high-rise steel structures and comprise the steps of acquiring cable tension data of a power transmission tower to be evaluated in real time, synchronously carrying out quick matching through a pre-constructed state evaluation sample database, and carrying out accurate prediction through a pre-constructed state evaluation model so as to comprehensively output an evaluation result. The database and the model are constructed in advance in the following way, based on finite element models of the sample tower, wind power action and inhaul cable tension time sequence data, a state label is generated by assimilation of Kalman filtering data, and then a state evaluation model is obtained through training, and the sample database is constructed. The method solves the problem of real-time quantitative evaluation of the power stability of the power transmission tower in the wind-driven force process, realizes the transition of the evaluation mode from static post-verification to dynamic real-time early warning, and provides core technical support for the precise safe operation and maintenance of the power transmission tower.

Inventors

YANG DEDONG
CHI CHAOFAN
ZHENG LI
WANG YU
CAO MEIGEN
YAN JUN
HUANG ZHIQING
KONG FANFANG
ZHANG YI
ZHANG RUOYU

Assignees

国网浙江省电力有限公司温州供电公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (10)

1. The power transmission tower state evaluation method based on the tension of the inhaul cable is characterized by comprising the following steps of: acquiring inhaul cable tension data of a power transmission tower to be evaluated in real time; matching the inhaul cable tension data with a pre-constructed state evaluation sample database, and outputting a first evaluation result according to a matching result; Inputting the tension data of the inhaul cable to a pre-constructed state evaluation model, outputting a second evaluation result, and integrating the first evaluation result and the second evaluation result to obtain a comprehensive evaluation result; The construction process of the state evaluation model comprises the following steps: Constructing a finite element model of a sample power transmission tower; Acquiring wind power action time sequence data acting on the sample power transmission tower and corresponding inhaul cable tension time sequence data; driving the finite element model by using the wind power action time sequence data to obtain a rod piece axial force sequence of the sample power transmission tower; correcting the rod piece axial force sequence by using the inhaul cable tension time sequence data as an observation value through a state estimation algorithm, and obtaining a corresponding power transmission tower stable state index based on the corrected rod piece axial force sequence; correlating the inhaul cable tension time sequence data with the power transmission tower stable state indexes to construct the state evaluation sample database; and training the state evaluation model based on the state evaluation sample database to obtain the state evaluation model.
2. The transmission tower state evaluation method based on the cable tension as set forth in claim 1, wherein the correcting the rod shaft force sequence by using the cable tension time sequence data as an observation value through a state estimation algorithm comprises: Based on the wind power action time sequence data and the inhaul cable tension time sequence data, constructing a state space model of the sample power transmission tower through subspace identification; and taking the rod piece axial force sequence as an initial value, and carrying out iterative updating on the initial value through Kalman filtering based on the state space model and the wind power action time sequence data to obtain a corrected rod piece axial force sequence, wherein the inhaul cable tension time sequence data is used as an observation value for filtering updating.
3. The transmission tower state evaluation method based on the cable tension as set forth in claim 1, wherein the obtaining the corresponding transmission tower steady state index based on the corrected rod piece axial force sequence includes: Performing spectrum analysis on the corrected rod piece axial force sequence, and extracting dominant frequency components as excitation frequencies; Calculating a dynamic instability region boundary corresponding to the excitation frequency based on a parameter excitation stability theory; And calculating the shortest distance from the excitation frequency to the boundary of the dynamic instability region to obtain a structural instability factor, and taking the structural instability factor as the steady state index of the power transmission tower.
4. The transmission tower state evaluation method based on the cable tension as set forth in claim 1, wherein the training the state evaluation model based on the state evaluation sample database to obtain the state evaluation model includes: dividing the state evaluation sample database into a training set and a verification set; constructing a long and short-term memory neural network as an initial model; inputting the inhaul cable tension sequence in the training set to the initial model to obtain a predicted value of the power transmission tower steady state index; The method comprises the steps of adjusting parameters of an initial model through a back propagation algorithm with the aim of minimizing errors between the predicted value and the power transmission tower steady state indexes corresponding to the training set; And stopping training when the evaluation result of the initial model on the verification set meets a preset condition, and taking the current initial model as the state evaluation model.
5. The transmission tower state evaluation method based on cable tension as set forth in claim 4, wherein the long-short-term memory neural network includes an attention mechanism, and the inputting the cable tension sequence in the training set to the initial model, to obtain the predicted value of the transmission tower steady state index, includes: calculating the association weight between time sequence features in the inhaul cable tension sequence through the attention mechanism; carrying out weighted fusion on the time sequence features according to the association weights to generate context vectors representing the sequence global information; And outputting the predicted value based on the context vector and the final hidden state of the long-short-term memory neural network.
6. A transmission tower state evaluation system based on cable tension, comprising: The acquisition module is used for acquiring the tension data of the inhaul cable of the power transmission tower to be evaluated in real time; the matching module is used for matching the inhaul cable tension data with a pre-constructed state evaluation sample database and outputting a first evaluation result according to a matching result; The evaluation module is used for inputting the tension data of the inhaul cable into a pre-constructed state evaluation model, outputting a second evaluation result, and integrating the first evaluation result and the second evaluation result to obtain a comprehensive evaluation result; wherein the evaluation module further comprises: the construction unit is used for constructing a finite element model of the sample power transmission tower; The data acquisition unit is used for acquiring wind power action time sequence data acting on the sample power transmission tower and corresponding inhaul cable tension time sequence data; The axial force generating unit is used for driving the finite element model according to the wind force action time sequence data to obtain a rod piece axial force sequence of the sample power transmission tower; the database construction unit is used for correcting the rod piece axial force sequence by using the inhaul cable tension time sequence data as an observation value through a state estimation algorithm, and obtaining a corresponding power transmission tower stable state index based on the corrected rod piece axial force sequence; the association unit is used for associating the inhaul cable tension time sequence data with the power transmission tower stable state indexes and constructing the state evaluation sample database; The training unit is used for training the state evaluation model based on the state evaluation sample database to obtain the state evaluation model.
7. The transmission tower status assessment system based on cable tension as recited in claim 6 wherein said database construction unit is further configured to: Based on the wind power action time sequence data and the inhaul cable tension time sequence data, constructing a state space model of the sample power transmission tower through subspace identification; and taking the rod piece axial force sequence as an initial value, and carrying out iterative updating on the initial value through Kalman filtering based on the state space model and the wind power action time sequence data to obtain a corrected rod piece axial force sequence, wherein the inhaul cable tension time sequence data is used as an observation value for filtering updating.
8. The transmission tower status assessment system based on cable tension as recited in claim 6 wherein said database construction unit is further configured to: Performing spectrum analysis on the corrected rod piece axial force sequence, and extracting dominant frequency components as excitation frequencies; Calculating a dynamic instability region boundary corresponding to the excitation frequency based on a parameter excitation stability theory; And calculating the shortest distance from the excitation frequency to the boundary of the dynamic instability region to obtain a structural instability factor, and taking the structural instability factor as the steady state index of the power transmission tower.
9. The transmission tower status assessment system based on cable tension as recited in claim 6 wherein said training unit is further configured to: dividing the state evaluation sample database into a training set and a verification set; constructing a long and short-term memory neural network as an initial model; inputting the inhaul cable tension sequence in the training set to the initial model to obtain a predicted value of the power transmission tower steady state index; The method comprises the steps of adjusting parameters of an initial model through a back propagation algorithm with the aim of minimizing errors between the predicted value and the power transmission tower steady state indexes corresponding to the training set; And stopping training when the evaluation result of the initial model on the verification set meets a preset condition, and taking the current initial model as the state evaluation model.
10. The transmission tower condition assessment system based on cable tension of claim 9, wherein the training unit is further configured to: calculating the association weight between time sequence features in the inhaul cable tension sequence through the attention mechanism; carrying out weighted fusion on the time sequence features according to the association weights to generate context vectors representing the sequence global information; And outputting the predicted value based on the context vector and the final hidden state of the long-short-term memory neural network.

Description

Power transmission tower state evaluation method and system based on inhaul cable tension Technical Field The invention relates to the technical field of wind vibration control of high-rise steel structures, in particular to a power transmission tower state evaluation method and system based on inhaul cable tension. Background The power transmission tower is used as a key node of the power transmission network and is exposed to complex and changeable meteorological environments for a long time, and the structural safety of the power transmission tower is directly related to the reliability and public safety of the power grid. Wind, which is the most common and extremely destructive natural environment factor, causes a power effect that is a main cause of structural failure of a power transmission tower and reverse tower accidents. Therefore, the real-time and accurate monitoring and evaluation of the structural state of the power transmission tower under the action of wind-driven force is a core technical challenge in the fields of disaster prevention, disaster reduction and intelligent operation and maintenance of the power transmission line. In the prior art, the assessment of the state of the power transmission tower under the action of wind power depends on post static assessment, namely, strictly follows the inherent logic of design-verification, and the core operation is to measure and qualify structural damage after wind disaster. The method leads to serious lag in evaluation timeliness, can not capture dynamic changes of stress states of the structure in real time in the wind action process, is difficult to early warn about impending instability risks, and can only take remedial measures after accidents, so that power interruption loss and structure repair cost caused by the serious lag are irrecoverable. Disclosure of Invention The invention provides a power transmission tower state evaluation method and system based on inhaul cable tension, which aim to solve the technical problem that real-time power instability risks of a power transmission tower under the action of wind-driven force are difficult to quantitatively early warn, so as to realize real-time, accurate and prospective evaluation of the safety state of the power transmission tower. In order to solve the technical problems, an embodiment of the present invention provides a transmission tower state evaluation method based on cable tension, including: acquiring inhaul cable tension data of a power transmission tower to be evaluated in real time; matching the inhaul cable tension data with a pre-constructed state evaluation sample database, and outputting a first evaluation result according to a matching result; Inputting the tension data of the inhaul cable to a pre-constructed state evaluation model, outputting a second evaluation result, and integrating the first evaluation result and the second evaluation result to obtain a comprehensive evaluation result; The construction process of the state evaluation model comprises the following steps: Constructing a finite element model of a sample power transmission tower; Acquiring wind power action time sequence data acting on the sample power transmission tower and corresponding inhaul cable tension time sequence data; driving the finite element model by using the wind power action time sequence data to obtain a rod piece axial force sequence of the sample power transmission tower; correcting the rod piece axial force sequence by using the inhaul cable tension time sequence data as an observation value through a state estimation algorithm, and obtaining a corresponding power transmission tower stable state index based on the corrected rod piece axial force sequence; correlating the inhaul cable tension time sequence data with the power transmission tower stable state indexes to construct the state evaluation sample database; and training the state evaluation model based on the state evaluation sample database to obtain the state evaluation model. As one preferable solution, the correcting the rod axial force sequence by using the cable tension time sequence data as an observation value through a state estimation algorithm includes: Based on the wind power action time sequence data and the inhaul cable tension time sequence data, constructing a state space model of the sample power transmission tower through subspace identification; and taking the rod piece axial force sequence as an initial value, and carrying out iterative updating on the initial value through Kalman filtering based on the state space model and the wind power action time sequence data to obtain a corrected rod piece axial force sequence, wherein the inhaul cable tension time sequence data is used as an observation value for filtering updating. As one preferable solution, the obtaining a corresponding power transmission tower steady state index based on the corrected rod member axial force sequence includes: Performing spectr