Increasingly the prevalence of electric vehicles (EVs) calls for an effective planning of charging stations by coupling transportation network (TN) with power distribution network (PDN). This paper conducts an interdisciplinary study on the coordinated planning strategy of EV charging stations and coupled traffic-electric network. A comprehensive model is proposed, which determines the optimal expansion strategies for both TN and PDN, including sites and sizes of new charging stations, charging spots, TN lanes and PDN lines. In TN, we use an unconstrained traffic assignment model (UTAM) to explicitly capture the steady-state distribution of traffic flows. In PDN, operating conditions are described by a linearized DistFlow. Moreover, EV charging demands and interdependency of TN and PDN are characterized by EV charging station location model (ECSLM). A mixed-integer linear programming (MILP) with UTAM constraints is formulated for the coordinated planning model, where a global optimal solution is obtained through a deterministic branch-and-bound method. Numerical experiments on a coupled traffic-electric network are conducted to validate the effectiveness of the proposed model and the solution method.

Line hardening refers to strengthening certain distribution lines that will not be subject to outages during extreme conditions. Line hardening provides a resilient solution in the case of major faults in a radial or meshed distribution system. We propose a robust optimal line hardening method coupled with multiple provisional microgrids to improve the distribution system resilience against worst N-k contingencies. A tri-level optimal model is considered with the objectives of minimizing the costs of line hardening and the operation of multiple islanded provisional microgrids, which could include the cost of load shedding in each provisional microgrid considering the worst N-k contingencies. Specifically, a two-stage robust optimization model is formulated and Benders decomposition (BD) algorithm coupled with iterative relaxation procedure (IRP) is developed to enable the tractable computation. In our work, the master problem determines line hardening strategies and the subproblem discovers the impact of the worst N-k contingencies corresponding to the optimal operation of reformed provisional microgrids. The second-stage max-min subproblem is solved efficiently using linearization techniques, duality theory and IRP. Our computational results for the IEEE distribution test systems validate the effectiveness of the proposed model and reveal that distributed generation is critical in increasing the resilience of distribution systems corresponding to the worst N-k contingencies in provisional microgrids.

Traffic lights play a critical role in mitigating traffic congestions, which can be energized by distributed generators (DGs) when power outages occur in urban areas. This paper studies the resilience enhancement strategy by line hardening and DG placement when outages occur in distribution lines and traffic lights in the coupled power distribution system and urban transportation system (PDS-UTS). A tri-level optimization model is formulated with a limited budget for line hardening and DG placement to minimize the cost of load shedding and aggregated vehicle travel time. The first level is to determine line hardening and DG placement strategies, the second level is to search for the worst case of faulted lines that would maximize load shedding and aggregated vehicle travel time, and the third level is to minimize the corresponding costs of load shedding and travel. In UTS, a dynamic user equilibrium (DUE) model is established in a cell transmission model (CTM) and solved by a linear complementarity approach. As the number of unavailable lines and traffic lights are definite in the inner-most level, the coupled PDS-UTS is considered as two decoupled systems. Accordingly, the tri-level model is converted into an equivalent bi-level model through Karush-Kuhn-Tucker (KKT) conditions, which is then solved by a greedy search method. Case studies corroborate the effectiveness of the proposed model and relevant solution method for the coupled PDS-UTS.

Constrained by low capacity and volatility, the rapid growth of distributed energy resources are obviously slowdown resulting in consumption difficulty and investment obstacle. As an effective integration and management technology, virtual power plant (VPP) becomes a suitable cornerstone of renewable energy future development. Based on current scientific research, this study intends to provide a detailed review of VPP from an internal perspective (e.g. energy resources’ integration and operation) to the external aspect including participation in electricity market. In accordance with market diversity, different bidding strategy optimisation problems of VPP are formulated and their corresponding mathematical solution methods are literately reviewed. To extract characteristics of VPP, a comparison between energy management techniques (e.g. VPP, microgrid, active distribution network, and load aggregator) is conducted, where advantages and defects of VPP are also concluded. Meanwhile, realistic deployments of VPP in European electricity markets are elaborated to verify its feasibility and applicability. To better accommodate future development of renewable energy, a flexible structure and effective management mechanism of VPP should be built leading to its technical innovation and management reform. Possessed with the threefold development orientation, VPP will undoubtedly improve the utilisation and management of renewable energy sources as a coordinated and operational entity.

Photovoltaic generation is playing a significant role in realizing the target of carbon neutralization. This article proposes a two-stage robust optimization model to assess the maximum hosting capacity of photovoltaic generation in power distribution network (PDN) considering the integration of soft open point (SOP) and electric vehicles (EVs). Convex relaxation method is applied to convert the two-stage robust optimization model to the form of mixed-integer second-order cone programming. A solving strategy based on the column-and-constraint generation algorithm is developed to solve the model, in which linearized difference-of-convex strategy is proposed to deal with the relaxation gap. Case studies are carried out on the modified IEEE 33-bus PDN integrated with SOP and EVs, and the results show the advantages of the proposed model and solving strategy.