In order to prevent behaviors of simultaneous commutation failure in the multi-infeed DC system

we proposed a dynamic VAR configuration method for reducing the risk of DC commutation failure to optimize the type

location and capacity of the dynamic reactive power compensation device. Firstly

the impedance matrix of the system nodes and the DC distance coupling index were calculated to identify the weak areas of reactive power compensation and the set of critical faults. The index of commutation failure duration shortening effect and the sensitivity index were calculated by time domain simulation. Therefore

the candidate nodes could be selected based on these indexes. Seocndly

after establishing the reactive power allocation optimization model of reducing the risk of DC commutation failure

we could obtain the Pareto optimal solution set by the MOEA/D. Finally

the close degrees of every decision schemes were calculated by the entropy weight method. And based on the calculation results

the relative superiorities of decision schemes were assessed which can help the operators to choose the effective compromise solutions. Calculation results of actual power grid show that the risk of simultaneous commutation failure of multi-infeed HVDC can be better reduced by simultaneously optimizing the type

location and capacity of the dynamic reactive power compensation devices. The results provide references for dynamic reactive power compensation of the multi-infeed DC system.