Multistage reliability-based expansion planning of ac distribution networksusing a mixed-integer linear programming model

A new mathematical model for the multistage distribution network expansion planning problem consideringreliability is proposed in this paper. Decisions related to substation and branch expansion are driven bythe minimization of the total cost, which comprises investment and operating costs including the impactof reliability. The proposed model features two main novelties. First, a set of novel algebraic expressions isdevised for a standard reliability index, namely the expected energy not supplied. As a result, the dependence ofreliability on network topology is explicitly and effectively cast in the mathematical formulation of the planningproblem at hand. In addition, the effect of the network is characterized by a computationally efficient piecewiselinear representation of the ac power flow model that takes into account both real and reactive power. Theresulting optimization problem is formulated as an instance of mixed-integer linear programming, whichprovides a suitable framework for the attainment of high-quality solutions with acceptable computational effortusing efficient off-the-shelf software with well-known convergence properties. The effectiveness of the proposedplanning methodology is empirically demonstrated by providing cheaper expansion plans that enhance systemreliability and by achieving better computational results as compared with state-of-the-art models.

Automated summary

This paper proposes a new mathematical model for the multistage distribution network expansion planning problem considering reliability. The model minimizes the total cost and introduces novel algebraic expressions for a reliability index, providing a framework for high-quality solutions with efficient computational effort.