BI-LEVEL OPTIMISATION OF VIRTUAL POWER PLANT CONSIDERING SOURCE-LOAD COORDINATED RESPONSE AND DYNAMIC PRICING

In order to guide the followers to actively participate in the main body operation，a comprehensive dynamic pricing mechanism is proposed in combination with the evaluation indices such as system carbon emissions and load interruption. This mechanism can dynamically adjust the electricity selling price to the lower load according to the information such as carbon emissions and load interruption，and reasonably guide the lower load to participate in the upper optimization，so as to achieve win-win results for multiple stakeholders at the upper and lower levels. At the same time，combined with the demand response characteristics of the electro-hydrothermal system and the electro-thermal load，a source-load coordinated response mechanism is proposed. The dynamic pricing mechanism is used as a link to realize the deep integration of the demand response characteristics of the upper-level electric-hydrogen thermoelectric system and the lower-level electric-heat load，improve the energy utilization efficiency of the energy side，and jointly realize the coordinated economic operation of the upper and lower layers. In addition，based on the electrical and thermal load dissatisfaction coefficient，the load energy dissatisfaction model is constructed and included in the system operation cost. Secondly，in order to deal with the uncertainty analysis of wind-solar power generation and load，a robust optimization method considering uncertain parameters is introduced to deal with the source-load uncertainty，so as to realize the flexible coordination between economy and robustness of the virtual power plant. At the same time，the influence of uncertain parameters on the operation of the virtual power plant is realized in depth. Finally，CPLEX-C&CG is used to solve the two-layer model iteratively，and the joint optimal solution of the system is selected according to the joint optimal discriminant mechanism. The results of the example verify the effectiveness of the model and method. © 2024 Science Press. All rights reserved.