Multi-stage planning of active distribution networks considering the co-optimization of operation strategies

Expansion planning for active distribution networks considering deployment of smart management technologies

“…(19) In order to demonstrate the accuracy of the linearisation to the power flow equations after attaining the optimal solution, we present the voltage magnitude difference between our results and those of the Newton-Raphson method. A comparison is also made between our method and that of the lossless DC power flow in [4]. Fig.…”

“…Mohtashami et al [3] present a multi-epoch DNP optimisation model in a distribution system with high penetration of DGs, and smart grid technologies are considered. Shen et al [4] propose a co-optimisation method to solve the expansion planning problem of an active distribution network (ADNΨ with energy storage systems (ESSsΨ by minimising the long-term investment costs and the short-term operating costs. A bi-level model for ADN expansion planning considering renewable energy generation is proposed in [5], wherein the upper level addresses the investment decision-making problem while the lower one optimises the demand level according to the profiles of electricity prices.…”

“…Besides, more and more expansion alternatives, such as VRs and capacitor banks, are incorporated into the multistage DNP to mitigate the voltage deviation problem [9]. On the other hand, the operation cost includes the maintenance cost, energy loss, operation cost of substations [2], DG curtailment cost and loadshedding cost [3], costs of electricity procurement from the upstream power system and DGs [4], and reliability cost [10]. It should be mentioned that the reliability assessment is typically considered in the operation stage in some existing publications.…”

“…The flexible topology is the key feature of the ADN through remote-control switches [4,16], which can effectively reduce the network loss and minimise the energy shortage through distribution network reconfiguration (DNRΨ. Cortes et al [17] offer the optimal design of a microgrid topology in ADN following an iterative procedure, wherein graph partitioning, integer programming, and performance index are applied for the optimal design.…”

“…However, the conditions for the exact conic relaxation is too strict for the DNP model as the direction of power flow can hardly be determined in advance [23]. In [4,11,12], a lossless DC power flow model is applied, which transformed the original DNP model into a MILP model. In our paper, some linearisation and approximation techniques are employed to transform the branchflow-based DNP model into a mixed-integer quadratic constrained programming (MIQCPΨ problem which is tractable for most offthe-shell solvers.…”