A resilient and carbon-aware virtual power plant coordination framework for low-carbon energy systems using multi-objective optimization

With the rapid deployment of distributed energy resources (DERs) such as photovoltaics and wind generators, traditional distribution networks are gradually developing into decentralized and resilient energy networks. In this regard, virtual power plants (VPPs) can help coordinate the operations of these distributed sources to achieve efficiency gains, reduced carbon emissions, and increased system resiliency. This study introduces a coordinated VPP scheduling framework for resilient and low-carbon distribution system operation, where modified ant lion optimization algorithm (MALOA) is employed as the optimization engine for solving the multi-objective coordination problem. The VPP is implemented on an IEEE 69-bus radial distribution system (RDS) for 24 h taking into account different weather scenarios, namely clear, cloudy, and windy days. Moreover, a multi-objective optimization approach with an adaptive weight function is proposed to optimize the operation cost, emission level, power losses, and resiliency penalty. The VPP also includes a contingency-aware microgrid creation process and a two-phase transactive energy trading scheme to ensure the reliable operation of the VPP when faced with faults. To efficiently coordinate DER scheduling, resilience-aware energy trading, and fault-driven microgrid operation under dynamic operating conditions, a problem-oriented MALOA-based optimization strategy is employed within the proposed VPP coordination framework. Simulation results show that the suggested approach lowers the operating cost to 28, 420 and the amount of CO 2 emission to 33. 2 tons while utilizing a maximum percentage of renewables which is 85%. The resilience index of the microgrids in case of fault events is estimated to be 0. 90 for multiple faults; in both cases, the amount of load curtailment is not more than 8. 5% for multiple faults. It can also be noticed that the suggested approach has outperformed other algorithms such as PSO, GA, GWO, and traditional ALO in terms of multi-objective function values.