Proceedings of PURPLE MOUNTAIN FORUM 2019-International Forum on Smart Grid Protection and Control pp 443-454 | Cite as
Risk Assessment Method for Smart Substation Secondary System Based on Deep Neural Network
- 322 Downloads
Abstract
In the risk assessment for the smart substations, the traditional Monte Carlo-based methods rely on prior distribution knowledge and cannot cover all the potential failure scenarios. In this paper, a risk assessment method based on the deep neural network for the smart substation secondary system is proposed. Firstly, a deep neural network established by the deep auto-encoders is proposed to quantitatively evaluate the operational risk of the smart substation. Secondly, the key indicators that affect the substation operations are hierarchically combed and refined, and are used as the inputs of the deep neural network. Finally, numerical simulation results from an actual smart substation show that compared with the traditional Monte Carlo-based assessment methods, the accuracy of the proposed method for assessing the operation states of the smart substation can be improved by 48.03% under the same iterations. In addition, the running time of the proposed method is less by 12.3% than the time of traditional method in the case with the same iterations. Hence, effectiveness and feasibility of the proposed method can be verified.
Keywords
Smart substations Secondary systems Risk assessment Deep neural network Deep auto-encodersNotes
Acknowledgements
This work was supported by Science and Technology Project of State Grid Chongqing Electric Power Company in 2018. The project name is “Integrated Operational Support Technology of Intelligent Substations Based on Total Service Data” (No. 2018#35).
References
- 1.Yang T, Rui Z, Zhang W et al (2017) On the modeling and analysis of communication traffic in intelligent electric power substations. IEEE Trans Power Delivery 32(3):1329–1338CrossRefGoogle Scholar
- 2.Wu HT, Jiao CQ, Cui X et al (2017) Transient electromagnetic disturbance induced on the ports of intelligent component of electronic instrument transformer due to switching operations in 500 kV GIS substations. IEEE Access 5(99):5104–5112CrossRefGoogle Scholar
- 3.Hengtian WU, Jiao C, Cui X et al (2017) Analysis and simulated experiment for port disturbance voltage due to switching operation in GIS substation. High Voltage Eng 43(10):3387–3395Google Scholar
- 4.Ghoneim SSM (2018) Intelligent prediction of transformer faults and severities based on dissolved gas analysis integrated with thermodynamics theory. IET Sci Meas Technol 12(3):388–394CrossRefGoogle Scholar
- 5.Lubbers N, Smith JS, Barros K (2018) Hierarchical modeling of molecular energies using a deep neural network. J Chem Phys 148(24):241715CrossRefGoogle Scholar
- 6.West MD, Labat I, Sternberg H et al (2018) Use of deep neural network ensembles to identify embryonic-fetal transition markers: repression of COX7A1 in embryonic and cancer cells. Oncotarget 9(8):7796–7811CrossRefGoogle Scholar
- 7.Chen B, Li C, Qin H et al (2018) Evaluation of typhoon resilience of distribution network considering grid reconstruction and disaster recovery. Autom Electr Power Syst 42(6):47–52 Google Scholar
- 8.Bains M, Warriner D, Behrendt K (2018) Primary and secondary care integration in delivery of value-based health-care systems. Br J Hosp Med 79(6):312–315CrossRefGoogle Scholar
- 9.Yin Q, Duan B, Shen M et al (2019) Intelligent diagnosis method for open-circuit fault of sub-modules in modular five-level inverter. Autom Electr Power Syst 43(01):162–170Google Scholar
- 10.Śniegocki M, Nowacka A, Smuczyński W et al (2018) Intramedullary spinal cord metastasis from malignant melanoma: a case report of a central nervous system secondary lesion occurred 15 years after the primary skin lesion resection. Postepy Dermatol Alergol 35(3):325CrossRefGoogle Scholar