An Integrated Risk Assessment Model for the Multi-Perspective Vulnerability of Distribution Networks under Multi-Source Heterogeneous Data Distributions
Junxian Huang,Chun Chen,Chenhao Sun,Yijia Cao,Yi An
DOI: https://doi.org/10.1016/j.ijepes.2023.109397
IF: 5.659
2023-01-01
International Journal of Electrical Power & Energy Systems
Abstract:The reliability of distribution network transmission corridors are vulnerable to different risk factors in both internal operation parameters and surrounding environmental conditions, and thus pinpoint evaluations of these impacting factors can further enhance their safety. To this end, this paper proposes an integrated assessment method, that is the multi-dimensional assessment with integrated fuzzy association (MDATFA). In this model, healthiness and importance indices of transmission lines are both defined to reflect the self-robustness and potential damages, respectively, and therefore to enable a two-dimensional assessment to incorporate the multisource heterogeneous inputs. Firstly, healthiness indices for environmental characteristics are analyzed. On the one hand, discrete input features are processed by the established association rule mining model with rare variables (ARMrv) to mine and diagnose rarely occurred elements, and their healthiness contribution values can then be calculated. On the other hand, continuous features are handled by the designed hierarchical probabilistic fuzzy systems (HPFS). The weight assignments of these two expert models are adjusted by a gating network. Secondly, the importance indices for electrical characteristics are calculated. Under the condition that electrical signal constraints are satisfied, the amount of load loss and lost users can be solved by the built distribution network fault recovery model which is according to heuristic rule and second-order cone (HRSO). Finally, the validity and robustness of this operational condition assessment model is verified by an empirical analysis in an actual 39-bus power distribution system.