There are problems of long idle time of demand-side energy storage accident backup and high cost of traditional thermal power units. Thus an economic optimization method considering standby risk of demand-side energy storage accident and thermal power unit deep peak shaving is proposed. First, a demand side ES risk quantification model and a standby peak adjustment economic model of ES accident are constructed by analyzing the weather state, load rate level and failure risk type. Secondly, the DPR economic model of fuel injection cost of a thermal power unit is built based on the analysis of unit damage and fuel demand during DPR. Finally, an economic optimization model of an ES accident standby auxiliary thermal power unit DPR based on demand side risk quantification is proposed. The particle swarm optimization algorithm is used to verify the calculation example in the modified IEEE 30-node system. The results show that the participation of ES accident standby in DPR can effectively improve the peak load adjustment capability and the utilization rate of ES accident standby. This work is supported by the Youth Fund of National Natural Science Foundation of China (No. 51707014) “Research on Transient Characteristics and Fault Traverse of Voltage Source Converter Connected in Series to Power Grid”.