Long-term risk assessment of complex distribution systems based on dynamic time-varying failure rates and multi-risk coupling mechanisms

With the growth of global energy demand and the transformation of energy structure, distribution systems face complex internal and external challenges such as climate change, equipment aging, integration of distributed energy resources, and the diversification of customer demands. Traditional risk assessment methods are no longer sufficient to meet the long-term risk assessment requirements of distribution systems. To address this, a long-term risk assessment method integrating dynamic failure rates and multi-risk coupling is proposed. This method explicitly considers dual uncertainties: the outer-layer environmental uncertainty and the inner-layer equipment failure uncertainty. Based on meteorological forecast data, the correlations between meteorological factors and load, photovoltaic (PV) generation, and insulation aging of distribution transformers are considered. Time series decomposition, empirical mode decomposition, and artificial neural networks are employed to achieve long-term prediction of the future operating environment of the distribution system. The outer-layer environmental uncertainty is processed using the Monte Carlo method, while the inner-layer equipment failure uncertainty is analyzed using the minimum path method to calculate the system loss of load risk. The proposed method also comprehensively assesses the risk of curtailed PV generation and the risk of equipment insulation aging. The effectiveness of the proposed risk assessment method is verified through the modified IEEE 123-node test system.