Data-driven voltage/var optimization control for active distribution network considering PV inverter reliability

•The XGBoost junction temperature calculation model is used to improve the junction temperature calculation efficiency and reduce the influence of IGBT parameter deviation on the junction temperature calculation accuracy.•A data-driven IGBT reliability evaluation method is proposed to realize the quantitative evaluation of IGBT reliability in PV inverter when photovoltaic power supply participates in reactive voltage regulation in distribution network.•The voltage/var optimization model of the distribution network considering the reliability of the PV inverter is established, which significantly improves the reliability, minimum lifetime and average lifetime of the PV inverter.•DDPG algorithm is used to realize the voltage/var optimization of distribution network considering the reliability of the PV inverter, which effectively improves the solution speed of the voltage/var optimization model and lays a foundation for realizing online optimal voltage/var control.