摘要
提出一种基于数据驱动的绝缘栅双极型晶体管(IGBT)可靠性评估方法,根据光伏电源有功功率、无功功率、太阳辐照度、环境温度定量分析光伏逆变器IGBT可靠性。该方法利用LightGBM机器学习模型刻画IGBT工况和IGBT结温之间的非线性映射关系,有效解决了IGBT可靠性评估中IGBT结温计算耗时长、依赖模型参数的问题。最后,基于IEEE 33节点配电系统对参与配电网无功调控的光伏逆变器IGBT可靠性进行量化评估分析。
In this paper,a data-driven IGBT reliability evaluation method is proposed to quantitatively evaluate IGBT reliability in PV inverters according to active power,reactive power,solar irradiance and ambient temperature.At the same time,LightGBM machine learning model is used to replace the traditional thermoelectric coupling model,which effectively improves the calculation efficiency of IGBT junction temperature and reduces the dependence of IGBT reliability evaluation results on IGBT model parameters.Finally,based on IEEE 33 node distribution system,the reliability of IGBT PV inverters participating in reactive power regulation of distribution network was evaluated quantitatively.
作者
张波
高远
李铁成
胡雪凯
王磊
Zhang Bo;Gao Yuan;Li Tiecheng;Hu Xuekai;Wang Lei(Key Laboratory of Distributed Energy Storage and Micro-grid of Hebei Province North China Electric Power University,Baoding 071003,China;School of Electrical Engineering,Southeast University,Nanjing 210096,China;Hebei Electric Power Research Institute,Shijiazhuang 050021,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2024年第6期296-302,共7页
Acta Energiae Solaris Sinica
基金
河北省自然科学基金(E2022502059)
国网河北省电力有限公司电力科学研究院科技项目(kj2022-021)。
关键词
太阳能
配电网
光伏逆变器
可靠性分析
LightGBM模型
结温
solar energy
distribution network
photovoltaic inverter
reliability analysis
LightGBM model
junction temperature
