基于改进QPSO-SVM的输电线路覆冰厚度预测

Prediction of Ice Thickness of Transmission Lines Based on Improved QPSO-SVM

冰雪灾害严重影响电网安全稳定运行,因此,对输电线路覆冰厚度进行预测可以有效地指导电网抗冰工作。利用灰色关联法分析温度、湿度、风速、降雨量和气压对线路覆冰厚度的影响,提出一种基于历史统计数据的输电线路覆冰厚度预测模型,采用改进量子粒子群算法(IQPSO)对支持向量机模型(SVM)参数惩罚因子c和核函数宽度δ进行寻优,结果表明:环境温度与覆冰厚度关联性最高,其次是降水量、风速、相对湿度,气压关联度最小,呈弱相关性;于同类型方法相比,IQPSO-SVM覆冰厚度预测模型预测精度较高,其平均绝对误差百分比为1.946%,均方根误差0.107 mm,验证了预测模型的有效性,具有一定的工程价值。

Ice and snow disasters seriously affect the safe and stable operation of the power grid,so predicting the ice thickness of transmission lines can effectively guide the anti-ice work of the power grid. The grey correlation method was used to analyze the effects of temperature,humidity,wind speed,rainfall and air pressure on the ice thickness of the line,and a prediction model of ice thickness of transmission lines based on historical statistics was proposed,and the improved quantum particle swarm algorithm(IQPSO) was used to optimize the parameter penalty factor c and the width of the kernel δ function of the support vector machine(SVM),and the results showed that the ambient temperature had the highest correlation with ice thickness,followed by precipitation,wind speed and relative humidity,and the air pressure correlation was the smallest and the weak correlation was carried out. Compared with the same type of method,the IQPSO-SVM ice thickness prediction model has higher prediction accuracy,with an average absolute error percentage of 1.946% and a root mean square error of 0.107 mm,which verifies the effectiveness of the prediction model and has certain engineering value.