Abstract：With the increasing complexity of distribution grids due to the integration of distributed power

sources，an improved matrix acceleration state enumeration method was proposed to enhance the computational

efficiency and accuracy of reliability assessment when integrating distributed energy into distribution grids. First，a

linear programming model was established to define reliability assessment indicators for distribution grids. Through

system simulation analysis，the weak links in the distribution grid were identified. Then，distributed power sources

were integrated into the weak links to effectively enhance the reliability of the distribution grid. Finally，to address

the issues of low computational efficiency and poor dynamic adaptability of traditional state enumeration methods

in complex distribution grids，an improved matrix-accelerated state enumeration method was proposed. The method reduced the dimensionality of the distribution grid matrix，incorporated LSTM wave prediction，reduced the computational variance of traditional distribution grid assessments，and improved computational accuracy. By

comparing reliability assessment methods for distribution grid systems with 76 nodes，126 nodes，377 nodes，and

957 nodes，the improved algorithm's assessment efficiency was effectively validated as significantly enhanced. In

the complex distribution grid with 957 nodes，the improved algorithm achieved nearly 94% increase in

computational efficiency compared to the Monte Carlo method.

        Key words：state enumeration method；identification of weak points；evaluate efficiency；calculate the

variance；power supply reliability

     Format Citation:肖丽莎，程江洲，鲍刚,等. 基于改进矩阵加速状态枚举法的配电网供电可靠性提升策略[J].电气传动,2026,56(07):81-90. XIAO Lisha，CHENG Jiangzhou，BAO Gang,et al. Power supply reliability improvement strategy for distribution network based on improved matrix accelerated state enumeration method [J].Electric Drive, 2026,56(07):81-90