Error prediction of optical current measurement device for DC pole line in converter station based on variational mode decomposition and extreme learning machine
LUO Qiang1, HUANG Yulei2, YAN Jun3, ZI Yuehua4, XU Tianqi2
1. Jiangsu LingChuang Electric Automation Co.,Ltd, Zhenjiang, Jiangsu 212000; 2. The Key Laboratory of Cyber-Physical Power System of Yunnan Universities, Yunnan Minzu University, Kunming 650504; 3. China Three Gorges Wuhan Science and Technology Innovation Park, Wuhan 430010; 4. Hua’neng Longkaikou Hydropower Co.,Ltd, Dali, Yunnan 671506
Abstract:With the rapid development of high voltage direct current (HVDC) transmission in China, optical principle based pole-to-pole DC measurement devices are widely used. Accurately predicting the trend of measurement errors is crucial for the operation and protection of HVDC transmission systems. In response to the problems of large prediction errors and low training efficiency in existing methods, a prediction method based on variational mode decomposition and extreme learning machine is proposed. The error time series of the measurement device is decomposed using variational mode decomposition, and then a particle swarm optimization algorithm is used to optimize the extreme learning machine for multi-step prediction of each mode. The predicted measurement error is obtained through reconstruction. Through comparison with multiple models, the superiority of the proposed method is verified.
罗强, 黄玉磊, 颜俊, 自越华, 徐天奇. 基于变分模态分解和极限学习机的换流站直流极线光学电流测量装置误差预测[J]. 电气技术, 2024, 25(11): 1-9.
LUO Qiang, HUANG Yulei, YAN Jun, ZI Yuehua, XU Tianqi. Error prediction of optical current measurement device for DC pole line in converter station based on variational mode decomposition and extreme learning machine. Electrical Engineering, 2024, 25(11): 1-9.