Abstract: The average reliability indices can not distinguish between the events of low probability but high damage and high probability low damage.In order to overcome these shortcomings,this paper proposes an extended risk analysis framework for the power system based on the partitioned multi-objective risk method.In the extended framework,the risk states were divided into three ranges: low-damage,moderate-damage and high-damage domains,and three conditional risk functions and conditional probabilities were established.On the basis of the capacity outage probability table,we proposed the risk assessment algorithm to calculate the conditional risk indices.Finally,the IEEE-RTS and the TH-RTS2000 were evaluated by the proposed framework and algorithm.The risk distribution and variation in different load levels were investigated and the effects of the damage-partitioned points were discussed.In the extended risk analysis framework,the analysts and decision makers can make trade-off between the average risk and the conditional risks of the high,medium,and low damage ranges,and they can comprehend the risk levels of the power system in a holistic and deep way.