Single-objective and multi-objective optimization for variance counterbalancing in stochastic learning

Triantali, Dimitra G.; Parsopoulos, Konstantinos E.; Lagaris, Isaac E.

doi:10.1016/j.asoc.2023.110331

Cited by 3 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…( 8), the error over time back propagation algorithm is applied to obtain the gradient value of each parameter in the model by the loss function, and the parameter value is updated with the gradient descent strategy. The gradient descent strategy chosen in this paper is Adam algorithm [36], and the parameter update formula is,…”

Section: ) Feature Fusion Processmentioning

confidence: 99%

Resilience Assessment of Urban Distribution Network Under Heavy Rain: A Knowledge- Informed Data-Driven Approach

Li¹,

Ma²,

Gao³

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Heavy rains pose a great threat to the reliable and secure power supply of urban distribution networks. A knowledge-informed data-driven resilience assessment approach is proposed to evaluate urban distribution networks' abilities to resist heavy rains. Firstly, the rainstorm waterlogging process is simulated to obtain the rainstorm intensity and rainfall process, providing the input for the data-driven model. Then, input variables are grouped guided by expert knowledge, and a dynamic and static data-driven model is constructed to predict the line outages based on historical data. Finally, the Monte Carlo simulation method integrated with the data-driven model is developed to assess the resilience of urban distribution networks and the number of line outages is selected as the evaluation metric. The effectiveness of the proposed method is sufficiently validated by the historical data of an urban distribution network.

show abstract

Section: ) Feature Fusion Processmentioning

confidence: 99%