Load fluctuation

high penetration of distributed generation and diversity of power electronic equipment make it more difficult to analyze the safe and stable operation of distribution network（DN）. In this paper

a critical node identification method of DN based on complex network theory（CNT） and utility risk theory is proposed

and the operation risk analysis of high penetration distributed generation accessing DN according to node criticality is carried out. The network dynamic characteristics and load sequence are taken into consideration

and the electric betweenness of lines at different times is calculated to realize the dynamic modeling of weighted topology for DN. Moreover

the security

economy and network structure stability are taken into consideration

and a multi-angle identification index system of critical nodes of DN is established. Then

the subjective and objective weights of the indexes are calculated based on the improved analytic hierarchy process（IANP） and improved entropy weight method（IEWM）

and the node criticality comprehensive evaluation and ranking of different time sections are realized by combining with the technique for order preference by similarity to ideal solution（TOPSIS）. Finally

a simulation model based on DN historical operation data and equipment parameters is established

then the simulation analysis of the high penetration photovoltaic（PV） power supply connected to DN is carried out to verify that the proposed method is effective in improving the voltage stability margin and reducing the branch load rate in DN

and thus the application of CNT in the urban energy internet can be promoted.