Heuristics for efficient transmission network expansion planning with load uncertainties

Simultaneous generation and network expansion planning in large‐scale power systems under exact AC power flow equations

“…From the viewpoint of optimization methods to provide optimal schemes in a computationally tractable manner, the previous studies can be classified as metaheuristic-and mathematical-based ones. The first class of references applies approaches like real genetic algorithm (GA) [8,9], decimal codification GA [47,48], non-dominated sorting genetic algorithm II (NSGA II) [17,18], particle swarm optimization (PSO) [10][11][12]29], improved discrete PSO [49,50], a combination of PSO and GA (PSO+GA) [27], differential evolution (DE) [13][14][15][16], DE-continuous population-based incremental learning (DE-PBILC) [24,25], artificial bee colony (ABC) [46], modified ABC [19][20][21], discrete ABC [52], improved discrete ABC [51], ant colony optimization algorithm for continuous domains (ACOR) [23], a combination of firefly algorithm and harmonic search algorithm (FFA+HAS) [39], adaptive tabu search (TS) [53], and hybridization of GA, TS, and artificial neural network (GA + TS + ANN) [28]. The latter one uses interior point method (IPM) [7], primal-dual interior-point method (PDIPM) [26], benders decomposition [34,38,40,44] or describes MINLP [22,36], mixed-integer linear programming (MILP) [30-33, 39, 41, 45, 52], and mixed-integer second-order cone programming (MISOCP) …”

“…Note that if new transformers are planned to be added to an existing substation or more than one transformer is planned for construction in a new substation, the corresponding investment cost can be assumed less than constructing a new transformer in a new substation. Therefore, if the total number of new and existing transformers in a substation is greater than one, the auxiliary variable Γℑ n will be set by (23). Then, disjunctive constraints ( 24)-( 29) reflect the investment cost of new transformers.…”

“…On the other hand, the NEP problem identifies the optimal number of circuits, voltage levels, timing, location, size, and technology of new network facilities, including transmission lines and power transformers within existing/candidate power stations, to supply the future load demand. It is noteworthy to mention that the inherent complexity of the expansion planning problems has led researchers to solve the GEP and NEP problems separately as GEP [2–4], transmission expansion planning (TEP) [5–35], coordinated GEP and TEP (G‐TEP) [36–44], and NEP [45–55]. The authors of [2–4] pay considerable attention to the prediction of uncertainties and the efficiency of the obtained generation expansion plans in a DC power flow‐based framework.…”

“…The majority of researches such as [2–55] have examined economic indices like planning and operation costs in TEP, NEP, and G‐TEP problems. However, without modelling GEP and NEP simultaneously (G‐NEP), there is no guarantee to obtain the most cost‐effective expansion plan in today's practical large‐scale power systems.…”