Against the backdrop of high penetration rates of renewable energy

the abrupt reduction in output caused by extreme weather conditions poses a significant challenge to the safe operation of the power system.Therefore

accurately assessing the depth of renewable energy output under extreme weather conditions is crucial for formulating operational strategies for the power system.This paper proposes a new method to evaluate renewable energy output using an improved deep temporal convolutional network(DeepTCN)approach.By designing a temporal convolutional neural network(CNN)architecture with dynamically weighted inputs

this method precisely quantifies and estimates the impact of extreme weather on renewable energy output

providing risk assessment information for formulating the operation mode of power systems under extreme weather conditions.Calculation results on the actual renewable energy output dataset of the power system show that

compared with conventional time series prediction methods

the proposed method can improve the normalized root mean square error

symmetric mean absolute percentage error

and mean absolute scaled error indicators by 1.2

0.1

and 0.22

respectively.Therefore

this approach enables a more accurate evaluation of renewable energy output under extreme weather conditions.