| The safe operation of power lines is an important factor to ensure social stability.The freezing disaster in extreme weather will cause the transmission lines to be covered with ice.The various accidents caused will not only cause irreversible damage to power facilities,but also increase the difficulty of inspections for power workers.In addition,it will bring inevitable economic losses to all power users,and it will also bring great inconvenience to the people’s lives.Grid companies attach great importance to this.To ensure the safety of power transmission in icy environments,the development of an optical fiber sensor-based transmission line ice coating monitoring system can effectively identify and warn accidents.In this paper,based on the meteorological data of heavily iced areas in northeast Yunnan of Yunnan Power Grid,the theoretical calculation model of ice thickness is studied with power lines as the object,and the data collected by optical fiber sensors are used to monitor the ice coating.In view of the cross-sensitivity of strain and temperature in the fiber tension and inclination sensors,as well as the influence of various factors in the working environment,the necessary temperature compensation for the measured value is required.Using temperature information and measurement information as the input of BP neural network can effectively compensate the temperature of the sensor,and realize the high-precision monitoring of the thickness of transmission line ice coating.The main research content of the paper is as follows:(1)The derivation of the equivalent thickness calculation model of transmission line ice coating.Based on the static model of the unloaded wire,the vertical plane,the wind deflection plane,and the plane with ice coating are analyzed mechanically in order,and the ice coating load and ice coating thickness are calculated.ANSYS finite element analysis software is used to simulate the force situation in different ice coating states in the tower-tower coupling system,which provides a theoretical basis for the placement of sensors.(2)Research on temperature compensation of optical fiber sensor in ice coating monitoring system.The fiber grating sensor used in the ice coating monitoring system will work together with strain and temperature during use,which will have a certain influence on the output result.The least square method and BP neural network algorithm can be used to effectively solve the problem of sensor temperature compensation.The ice thickness after compensation is input to the data of a transmission line ice monitoring system of Yunnan Power Grid,and the actual ice thickness and theoretical calculation of ice coating Thickness is analyzed to determine the state of ice coating.(3)Research on temperature compensation algorithm of BP neural network optimized by SWPA.In order to improve the optimization ability and convergence speed of the neural network algorithm,the wolf swarm algorithm based on adaptive step size and the whale algorithm based on dynamic approximation are given,and the performance of the algorithm is verified by the test function.The swarm intelligence algorithm is used to optimize the weights and thresholds of the BP neural network algorithm,and the results are compared with different algorithms.The experimental results show that SWPA-BPNN is an effective temperature compensation method. |
PDF Full Text Request