Reliability assessment for distribution optimization models: A non-simulation-based linear programming approach

Abstract

The advent of smart grids and active distribution networks has boosted the relevance of reliability in the operation and planning of distribution systems. As is customary, reliability is assessed analytically through several standard indices. Unfortunately, analytical reliability assessment relies on simulation, thereby requiring the use of inexact heuristic-or metaheuristicbased solution methods to operate and plan distribution systems when economic and reliability criteria are jointly considered. In order to overcome this shortcoming, this paper presents a new optimization-based approach to compute the standard network-dependent reliability indices that are widely used in reliability-constrained distribution optimization models. As a major salient feature over the conventional simulation-based method, reliability indices are equivalently determined by an efficient approach based on linear programming where the network topology is explicitly represented by decision variables of the optimization process. The proposed approach has been tested on several benchmarks including a 1080-node system. Numerical simulations show that the proposed approach yields the same results as the conventional algorithm. Moreover, the moderate computational effort is suitable for the subsequent integration of the proposed equivalent formulation in reliability-constrained optimization models for distribution operation and planning. Such successful numerical experience backs the potential of the proposed formulation to enable the use of sound techniques different from the available heuristics and metaheuristics to solve reliability-constrained operational and planning optimization models for distribution systems.Index Terms-Analytical reliability assessment, distribution optimization models, linear programming, non-simulation-based approach. He is currently a Full Professor of electrical engineering at the Universidad de Castilla-La Mancha. His research interests include operations, planning, and economics of power systems, as well as optimization.

Linear Formulations for Topology-Variable-Based Distribution System Reliability Assessment Considering Switching Interruptions

“…In other words, López et al's approach neglects switching interruptions, i.e., those with out-of-service duration equal to the switching time associated with the isolation of the faulty portion of the network. In [27], a pioneering model is proposed to derive a linear formulation for the calculation of reliability indices of radial systems while considering switching interruptions. This linear model can be incorporated into various distribution system optimization problems.…”

“…This linear model can be incorporated into various distribution system optimization problems. In [28], the model described in [27] is applied to distribution system expansion planning. However, although this method is capable of calculating all load-node and system-level reliability indices, it can considerably increase the dimension (number of decision variables and constraints) of the resulting optimization model.…”

“…In order to overcome this shortcoming, an enhanced algebraic approach is proposed in [29] for radial grids. Similar to [27], the technique proposed in [29] relies on the calculation of load-node indices to eventually compute system-oriented metrics. This can cause various challenges for modeling the impact of distributed generating units on reliability indices.…”

“…Secondly, additional constraints are incorporated to characterize the behavior of the healthy portion of the system upstream of the fault. Moreover, it should be noted that, unlike the formulation described in [27] and subsequently applied in [28], the proposed model does not require the computationally expensive consideration of system operational constraints under every contingency, which is beneficial for practical implementation purposes.…”

“…First, switching interruptions are considered, thereby significantly improving the accuracy of reliability indices upon those provided by [26] and [30]. In addition, the prespecification of a particular radial operation condition for mesh-designed grids is not required, unlike [27] and [29]. Note also that both modeling advantages are achieved in a computationally efficient way as the dimensionality issue of [27] and [28] is not featured.…”