Research On Icing Of Transmission Lines Based On Improved Quantum Particle Swarm Algorithm

Ice and snow disasters cause damage to transmission lines and towers,resulting in power supply interruptions in some areas and seriously affecting the safe and stable operation of power grids.Therefore,prediction of transmission line ice thickness can effectively guide the anti-icing work of power grids.In this paper,based on the analysis of line ice cover mechanism,a prediction model is established using machine learning intelligent algorithm,and the influence of different factors on line ice cover is analysed by grey correlation method,and a line ice cover thickness prediction model based on improved quantum particle swarm algorithm optimised support vector machine(IQPSO-SVM)is proposed.The main research elements of this paper are as follows.1)Describes the meteorological conditions and growth process of ice cover formation,summarises the three typical forms and hazards of line ice cover,takes into account the meteorological,topographical and line itself factors affecting line ice cover,and introduces two types of classical ice cover models.2)Multilayer perceptron(MLP)neural network and support vector machines(SVM)were used to establish a transmission line ice thickness prediction model;through simulation experiments,it is concluded that the regression prediction model established by SVM is more easily converged and less error than the MLP neural network prediction method.It is found that the regression prediction model established by SVM is more convergent and less error-prone than the MLP neural network prediction method,which is more suitable for transmission line ice thickness prediction.In the process of prediction,the penalty parameters and kernel functions of the SVM have a great influence on the accuracy of the results,and the Particle Swarm Optimization(PSO)algorithm is used to optimize the SVM parameters to reduce the relative prediction errors.3)In view of the large influence of meteorological factors on the prediction accuracy of the transmission line ice cover thickness model,the grey correlation analysis method is used to correlate the factors affecting the equivalent thickness of ice cover and obtain the strong correlation factors affecting line ice cover.The IQPSO-SVM line ice thickness prediction model is proposed to address the problem that the particle swarm algorithm tends to be limited to local optima in the iterative process,and the model is predicted through actual line ice coverage history data.The experimental results show that compared with similar methods,the IQPSO-SVM ice cover thickness prediction model proposed in this paper has a higher prediction accuracy,with an average absolute error percentage of 1.946% and a root mean square error of 0.107 mm,which has some engineering value.