Integrated resource expansion planning of wind integrated power systems considering demand response programmes

Abstract

Here, an optimisation framework is proposed for integrated resource expansion planning (IREP) including conventional generation units, wind generation units, and transmission lines while taking into account the role of demand response program (DRP) aggregators. This problem is a bi-level optimisation problem. In the upper-level problem, the objective function is to maximise the profit for each resource, that is, the generation company (GENCO), wind generation company (WINDCO), and transmission company (TRANSCO) as well as the DRP aggregator. Also, the lower-level problem considers a market model with the participation of private GENCOs, WINDCOs, TRANSCOs, and DRP aggregators. The lower-level problem minimises energy cost subject to AC power flow (PF) equations, power network limitations, pollution constraint, GENCO, WINDCO, TRANSCO constraints, and technical limitations of DRPs. Here, the lower-level problem is non-linear and non-convex. Accordingly, to facilitate the solution of the proposed bi-level optimisation problem, a linear model is proposed. Then, the proposed bilevel optimisation problem is converted into an integrated single-level one using the Karush-Kuhn-Tucker (KKT) conditions. Eventually, a mixed-integer linear programming (MILP) model is proposed. The proposed method is applied to the IEEE 6-bus and the IEEE 30-bus test systems, and finally, the capabilities of the proposed scheme are evaluated. Nomenclature Indices and sets (n/j), i, h, y, l, k Indices of buses, generations, hours, years, linearisation segments of line power flow terms and circular constraints, respectively. φ n , φ i , φ h , φ y , φ l , φ k , Sets of buses, generations, hours, years, linearisation segments of line power flow terms and circular constraints, respectively. Ω n Set of buses connected to bus n.

Bi‐level power management strategy in harmonic‐polluted active distribution network including virtual power plants

“…• To obtain a single level model for a multi-level problem by the KKT or duality methods, the constraints of lower level problems should be in a linear form [36].…”

“…The proposed PMS strategy modelled by (1)-(37) includes a non-convex NLP model due to non-linear equations of VDF, i.e. (4), (5), (10), (11), (14), (17), (22), (25), (28), (31), and (36), and the non-convex equations of (4) and (5) [25]. Hence, this model may obtain locally optimal solutions, which needs high computational time arising from non-convex constraints and numerical solving methods such as the Newton-Raphson approach [25].…”

“…subject to is the objective function of (58) so that min -f 2 in (63) is the same with max f 2 in (58), and g 1 -g 2 and g 3 -g 4 are related to constraints (57) and (59), respectively. It should be noted that the problem (60)-(65) must be expressed as an integrated model to solve this method using tools such as CPLEX solver in optimization software [36]. To cope with this issue, this paper uses the KKT conditions to convert the proposed bi-level optimization formulation to a single-level model [36].…”

Security-Constrained Optimal Protection Coordination for Dual-Setting Digital Directional Overcurrent Relays in the Distribution Network Including Non-Renewable/Renewable Synchronous Distributed Generation

“…(18)-(20), respectively 21 . Constraints (21) and (22) represent the corresponding equations with the capability curve of the SDG, except for its renewable type such as wind turbine system, which its generation active and reactive powers limits are presented in Eqs. (21) and (22), respectively 22,23 .…”

Two‐stage coupling of coordinated energy management and self‐healing strategies in a smart distribution network considering coupling neighbouring microgrids according to autonomous decentralized restoration technique

“…The objective function equation of the first stage problem is presented in Equation (1) that is applied to minimize the SDN expected energy cost of the upstream network 25‐28 . Hence, it has been predicted that the local power sources supply the SDN loads.…”