Search

This paper proposes a new approach to use adaptive neuro-fuzzy inference systems (ANFIS) for approximating the system security boundary (SB) to a differentiable function. The novelty of the proposal is that the representation of SB, which is generated by ANFIS, is used as the security constraint of an optimal power flow (OPF). The combination of the differentiable function and the constraint of the OPF will make the system reach its best economic benefit under the basic require of system security. The effectiveness of the proposed ANFIS method, as well as the feasibility of using the approach to analyze a layering and zoning network, is demonstrated through testing and simulation using the IEEE 118 bus test system. Results show that the proposed method can be used to the dispatch optimization in layering and zoning system, and is a better approach in speed and precision. [ABSTRACT FROM PUBLISHER]